0.14 pyladiescon-2025 2025-12-05 2025-12-07 3 00:05 https://pretalx.com https://pretalx.com/media/pyladiescon-2025/img/PyLadiesCon_2024_Slides_2_Q2wkdel.png UTC Main Stream Message from Organizers 2025-12-05T16:30:00+00:00 16:30 00:15 Welcome to the 2025 version of PyLadiesCon, we are happy to have you! Join Mariatta and Maria José that will start the conference! pyladiescon-2025-84079-conference-opening Maria Jose Molina ContrerasMariatta en false https://pretalx.com/pyladiescon-2025/talk/SXMCMY/ https://pretalx.com/pyladiescon-2025/talk/SXMCMY/feedback/ Main Stream Sponsors's talk 2025-12-05T16:45:00+00:00 16:45 00:05 Join our Champion sponsor for a quick introduction before our first Keynote! pyladiescon-2025-84096-champion-sponsor-intro Abby Mitchell en false https://pretalx.com/pyladiescon-2025/talk/9YAHXS/ https://pretalx.com/pyladiescon-2025/talk/9YAHXS/feedback/ Main Stream Keynote 2025-12-05T16:50:00+00:00 16:50 00:30 Being a leader is to guide people on a way, give advice and inspire folks in the direction we think is the right one. It's hard to be a good leader, people and yourself have expectations. You can't be fully ready for a leadership role. In this talk, you will learn the courage to lead imperfectly with my own example. pyladiescon-2025-84697-the-courage-to-lead-imperfectly Sarah Abderemane en false https://pretalx.com/pyladiescon-2025/talk/NF9DKL/ https://pretalx.com/pyladiescon-2025/talk/NF9DKL/feedback/ Main Stream Sponsors's talk 2025-12-05T17:20:00+00:00 17:20 00:15 Meta is proud to sponsor PyLadiesCon and to share our commitment to supporting the Python ecosystem through impactful open-source initiatives. In this talk, we’ll spotlight two major projects: Pyrefly: a fast, open-source Python type checker and language server written in Rust, designed to catch type-related errors before runtime and provide powerful IDE integration Free Threaded Python: a new addition to CPython (reached supported status earlier this year) that enables users to remove the Global Interpreter Lock (GIL) to enable true parallel execution of Python code across multiple threads, significantly improving multithreading performance. Join us to learn more about these initiatives, how to use them and how to get involved! pyladiescon-2025-84396-meta-open-source-highlights-pyrefly-and-free-threaded-python Abby Mitchell en false https://pretalx.com/pyladiescon-2025/talk/DF93G8/ https://pretalx.com/pyladiescon-2025/talk/DF93G8/feedback/ Main Stream Panel 2025-12-05T17:35:00+00:00 17:35 00:30 There is wonderful diversity within PyLadies, especially in the ways our members grow and make an impact in tech. Some have turned their Python contributions, whether open-source projects, community initiatives, or side projects, into real products, services, or meaningful outcomes. In this session, we will hear their inspiring stories and explore the journeys from contributor to creator, innovator, or community leader. Panelists will share lessons learned along the way, including practical insights for turning ideas into action, navigating challenges, and discovering new opportunities. Attendees will leave motivated, inspired, and with a deeper appreciation for the creativity, dedication, and impact within the PyLadies community. pyladiescon-2025-84086-from-contributor-to-founder-turning-python-projects-into-products Carol WillingSheenaInessa PawsonDeborah HanusLeah Wasser en false From Contributor to Founder https://pretalx.com/pyladiescon-2025/talk/DXBDRB/ https://pretalx.com/pyladiescon-2025/talk/DXBDRB/feedback/ Main Stream Talk 2025-12-05T18:10:00+00:00 18:10 00:20 An astonishing 96.3% of websites have accessibility issues. Turning this number around is not just a job for frontend developers. Python developers can help too! In this talk, I'll introduce you to the Web Content Accessibility Guidelines and show you practical ways to incorporate accessibility into your Python web development. I'll walk through examples created with a Python-powered content management system called Wagtail CMS. By the end of this talk, you'll learn the key ways Python developers can enhance web accessibility and you'll leave with a hands-on tutorial to build your own more accessible website. pyladiescon-2025-80320-building-more-accessible-python-powered-websites Meagen Voss en This talk is for Python enthusiasts who want a solid introduction to best practices in web accessibility. While attendees may find it helpful to have some experience with Django or Wagtail before the talk, anyone with some basic knowledge of Python syntax and functions will be able to follow along. This talk will start with a brief introduction to key web accessibility concepts and a Python-based CMS called Wagtail. Then I'll shift to three practical examples of how attendees can build accessibility best practices into their websites. I'll be using a blog built with Wagtail CMS as an example, but most of the broad concepts I'll cover can apply to Django and other Python web frameworks as well. At the end of the talk, I'll share a longer tutorial and other resources that attendees can use to extend their learning and dig more deeply into coding a website project with web accessibility as a focus. false https://pretalx.com/pyladiescon-2025/talk/NLGURP/ https://pretalx.com/pyladiescon-2025/talk/NLGURP/feedback/ Main Stream Talk 2025-12-05T18:30:00+00:00 18:30 00:20 La operación de drones en entornos complejos y peligrosos, como inspecciones industriales en interiores, presenta un alto riesgo de colisiones y requiere una gran destreza por parte del piloto. ¿Y si pudiéramos darle al piloto un copiloto de inteligencia artificial que anticipe errores y garantice la seguridad? En esta charla, exploraremos cómo desarrollar un sistema de teleoperación asistida para cuadricópteros utilizando Python, simulación y aprendizaje por refuerzo profundo (Deep RL). Partiremos desde cero, configurando un entorno de simulación fotorrealista con Webots para entrenar un agente de IA. Veremos cómo, usando librerías como stable-baselines3 e imitation, podemos enseñar a un dron a navegar de forma óptima y, lo más interesante, cómo usar ese conocimiento para asistir a un piloto humano. Profundizaremos en el concepto de "autonomía compartida", donde el sistema de IA no reemplaza al humano, sino que colabora con él. Mostraremos cómo simulamos pilotos con comportamientos "humanos" (con retrasos o errores) para entrenar un copiloto que interviene solo cuando es necesario, corrigiendo trayectorias peligrosas y previniendo accidentes. Al final de esta sesión, no solo entenderás los fundamentos para aplicar aprendizaje por refuerzo en robótica, sino que también verás un caso práctico de cómo Python se convierte en el puente entre una idea compleja de IA y una aplicación funcional con impacto en el mundo real. Descubrirás un flujo de trabajo completo para prototipar, entrenar y evaluar sistemas robóticos inteligentes de manera segura y eficiente, abriendo la puerta a innumerables aplicaciones industriales. pyladiescon-2025-80325-entrenando-un-copiloto-ia-para-drones-de-la-simulacion-en-python-al-vuelo-asistido Ana Laura Diedrichs es Los vehículos aéreos no tripulados (UAVs), o drones, han revolucionado industrias como la construcción, la seguridad y el mantenimiento de infraestructuras. Sin embargo, su mayor potencial se encuentra a menudo en los entornos más desafiantes: interiores de edificios, espacios confinados o áreas con obstáculos imprevistos. En estos escenarios, un error de cálculo del piloto puede resultar en una colisión costosa, daños al equipo o el fracaso de una misión crítica. La carga cognitiva para un operador humano es inmensa. Esta charla aborda una pregunta fundamental: ¿cómo podemos usar la inteligencia artificial para hacer que estas operaciones sean más seguras y eficientes, sin eliminar la intuición y el control del piloto humano? La Solución: un Copiloto de IA entrenado en un mundo virtual En esta presentación, te guiaré a través del proceso de creación de un "copiloto" de IA para un cuadricóptero Crazyflie. La clave de nuestro enfoque es la simulación. Antes de arriesgar un dron real, construimos un gemelo digital de nuestro entorno en Webots, un potente simulador de robótica de código abierto. Usando Python como nuestro lenguaje principal, integramos el simulador con Gymnasium, el estándar de la industria para entornos de aprendizaje por refuerzo. El Stack Tecnológico y Metodología 1) Entrenando al Piloto Perfecto: El primer paso fue entrenar un agente de IA para que se convirtiera en un piloto "óptimo". Utilizando la librería stable-baselines3, implementamos algoritmos de aprendizaje por refuerzo profundo (como PPO) para enseñarle al dron a navegar por el escenario simulado, evitando obstáculos y alcanzando su objetivo de la manera más eficiente. 2) Aprendiendo de los Humanos (o casi): Para que nuestro copiloto fuera realmente útil, necesitaba entender cómo ayudar a un piloto humano, no a otra IA. Aquí es donde entra en juego el aprendizaje por imitación. Con la librería imitation, utilizamos trayectorias generadas por un operador humano para refinar la política de nuestro agente, logrando un comportamiento más natural y menos "robótico". 3) Simulando la Imperfección Humana: El verdadero desafío era probar la asistencia. Para ello, creamos pilotos "subóptimos" programáticamente: un "piloto lento" que a veces tardaba en reaccionar y un "piloto ruidoso" que ocasionalmente cometía errores. Estos agentes simulados nos permitieron probar la eficacia de nuestro copiloto en un entorno controlado. 4) El Copiloto en Acción (Autonomía Compartida): El corazón de nuestro sistema es un modelo de autonomía compartida. El copiloto de IA recibe tanto las lecturas de los sensores del dron como las órdenes del piloto humano. Usando una función de calidad, evalúa si la acción del piloto es segura. Si la acción es segura, se ejecuta sin cambios. Si la acción podría llevar a una colisión, el copiloto interviene, eligiendo la acción segura más cercana a la intención original del piloto. Este equilibrio es crucial: asiste sin ser intrusivo. ¿Qué Aprenderá la Audiencia? Esta charla está diseñada para brindar inspiración práctica y conceptual. Al finalizar, los asistentes se llevarán: * Un entendimiento claro de cómo la simulación acelera el desarrollo en robótica e IA, reduciendo costos y riesgos. * Un mapa de ruta sobre cómo usar el ecosistema de Python (Webots, stable-baselines3, imitation, Gymnasium) para aplicar aprendizaje por refuerzo a problemas del mundo real. * Una introducción práctica al concepto de autonomía compartida y cómo puede ser más efectivo que la autonomía total en ciertas aplicaciones. * Insights / ideas sobre los desafíos encontrados y las lecciones aprendidas durante el proyecto, desde el diseño de la función de recompensa hasta la evaluación del rendimiento. La inspiración para aplicar estas técnicas en sus propios proyectos, ya sea en robótica, videojuegos, finanzas o cualquier otro campo donde se tomen decisiones secuenciales. Esta no es solo una charla sobre drones; es una historia sobre la colaboración hombre-máquina, facilitada por el poder y la flexibilidad del ecosistema de Python. false https://pretalx.com/pyladiescon-2025/talk/9UC39C/ https://pretalx.com/pyladiescon-2025/talk/9UC39C/feedback/ Main Stream Talk 2025-12-05T18:50:00+00:00 18:50 00:20 Large Language Models are trained to produce helpful, safe answers—but what happens when someone tries to make them misbehave? Malicious users can manipulate prompts to generate unsafe content like hate speech or violent instructions. Every time we add an LLM to an app, we open up that risk. In this talk, I’ll show how to use Python to red-team an LLM-powered app: simulating hundreds of bad actors to see how the system holds up. We’ll explore public datasets of adversarial prompts, and use the open-source pyrit package to obfuscate attacks with strategies like base-64 encoding and Caesar cipher. Finally, we’ll evaluate which attacks succeed, using another LLM to score the results. Attendees will walk away with a practical understanding of how to stress-test their own apps and a toolkit for keeping them safer against trolls. pyladiescon-2025-80310-beware-the-trolls-red-teaming-for-llm-powered-apps Pamela Fox en false https://pretalx.com/pyladiescon-2025/talk/EPHCZC/ https://pretalx.com/pyladiescon-2025/talk/EPHCZC/feedback/ Main Stream Sponsors's talk 2025-12-05T19:10:00+00:00 19:10 00:15 Managing Conda environments is essential for modern Python development, but it can be challenging. This is especially true for newcomers juggling long commands, complex dependencies, and the fear of breaking their installation. Gator offers a friendlier, visual way to create, clone, inspect, and manage environments, available both inside JupyterLab and as a standalone app launched with `gator`. This talk introduces Gator, explains how it fits into Anaconda Toolbox, and gives an overview of its architecture: a shared React UI, a lightweight Jupyter Server backend, and REST endpoints that delegate to Conda or Mamba. I’ll also cover the recent work to revive and modernize Gator for JupyterLab 4, including UI updates and improved workflows. Finally, we’ll look at what’s next: performance improvements, richer environment metadata, and deeper integration with other Toolbox features like Quick-Start Environments and the Anaconda Assistant. Whether you're new to Conda or a power user, Gator makes environment management smoother and more approachable. pyladiescon-2025-85796-gator-a-handy-ui-for-managing-conda-environments Rosio G Reyes en A friendly introduction to Gator, an open-source UI for managing Conda/Mamba environments. Learn how it works, how it was modernized for JupyterLab 4, how it integrates into Anaconda Toolbox, and what’s next for simplifying Python environment management. false https://pretalx.com/pyladiescon-2025/talk/WBDJAK/ https://pretalx.com/pyladiescon-2025/talk/WBDJAK/feedback/ Main Stream Talk 2025-12-05T19:30:00+00:00 19:30 00:20 Are you tired of doing the same tedious tasks over and over? This session is your opportunity to turn a computer from a chore into a collaborator. Geared toward beginners, this talk will be a practical, hands-on guide to using Python to automate the mundane and free up your time for more impactful work. Through live coding, we’ll tackle common annoyances like organizing messy file folders, sending routine emails, and even interacting with websites. We’ll cover the basic tools you need to get started and a few simple principles for making your automation scripts reliable. You’ll leave with the knowledge and confidence to build your own "personal assistant" scripts, empowering you to work smarter, not harder. Join us and discover the power of automation to reclaim your time and energy. pyladiescon-2025-80066-python-for-automation-getting-rid-of-repetitive-tasks-and-reclaiming-your-time The Joy of Tech en This talk is a hands-on introduction to the empowering world of automation using Python. The session is structured to get attendees quickly from concept to code, focusing on practical, real-world examples that can be immediately applied to their daily lives. The presentation will be fast-paced and demo-heavy, ensuring a dynamic and engaging experience. We'll start by building a compelling case for automation by exploring how a few lines of code can solve common, frustrating problems, from a cluttered Downloads folder to sending repetitive messages. The audience will see how these simple scripts reduce errors and significantly boost productivity. Next, we'll dive into the code. The presentation will cover: 1. File System Automation: Using built-in modules like os and pathlib, we will live-code a script that automatically renames and organizes files into logical folders based on type or date, eliminating the need for manual cleanup. 2. Web and Email Automation: We’ll explore how to connect to the outside world. I will demonstrate how to send an automated email with smtplib and how to interact with a basic web form using a library like Selenium. This section will highlight how to automate communications and information gathering. 3. Making It Run on Its Own: The final code demonstration will show how to use a simple scheduler, such as the schedule library, to make our scripts run automatically, turning a one-time script into a persistent solution. The talk will wrap up with a discussion of best practices, including simple error handling and the importance of clear, readable code. The ultimate goal is for every attendee to feel empowered to identify their own repetitive tasks and begin their own automation journey, recognizing that with Python, they have the tools to make their computer work for them. false Python pathlib Documentation Python smtplib Documentation Selenium Python Documentation https://pretalx.com/pyladiescon-2025/talk/7BB9MZ/ https://pretalx.com/pyladiescon-2025/talk/7BB9MZ/feedback/ Main Stream Talk 2025-12-05T19:50:00+00:00 19:50 00:20 Esta palestra é destinada para ser uma introdução ao PySpark, a API Python para executar o Apache Spark, com foco na aplicação em engenharia de dados. Serão apresentados os principais conceitos, como processamento distribuído, criação de DataFrames e operações de transformação. O objetivo é mostrar e simplificar o PySpark para iniciantes na área de Engenharia de Dados. pyladiescon-2025-80333-introducao-ao-pyspark-para-engenharia-de-dados Vitoria Rodrigues pt O PySpark é hoje uma das ferramentas mais utilizadas em engenharia de dados para lidar com grandes volumes de dados de maneira escalável e eficiente. Os participantes aprenderão os seguintes tópicos: - A diferença entre processamento local e distribuído; - Criação e manipulação de DataFrames no PySpark; - Exemplos de transformações comuns no dia a dia da engenharia de dados. A apresentação é voltada para iniciantes que desejam uma oportunidade na área de Engenharia de Dados. Ao final, os participantes terão uma visão clara de como o PySpark é utilizado no dia a dia de uma pessoa Engenheira de Dados. false https://pretalx.com/pyladiescon-2025/talk/MGAN8V/ https://pretalx.com/pyladiescon-2025/talk/MGAN8V/feedback/ Main Stream Panel 2025-12-05T20:10:00+00:00 20:10 00:35 Sponsor Panel: Tech industry and the job market pyladiescon-2025-83687-sponsor-panel-tech-industry-and-the-job-market FrankieParul GuptaZuotian TatumJessica Hedges en Sponsor Panel: Tech industry and the job market false Tech industry and the job market https://pretalx.com/pyladiescon-2025/talk/T8SZDC/ https://pretalx.com/pyladiescon-2025/talk/T8SZDC/feedback/ Main Stream Talk 2025-12-05T21:15:00+00:00 21:15 00:20 Communication has long been a necessary skill for data professionals, but most folks stop with simply data visualization and data storytelling. As a former marketing and PR manager, my seemingly “unrelated” experience has been a huge asset to my career in analytics and data science. We all know how to craft a story with data and create a compelling visualization. But do you know how to craft the story of the value of your team and build trust? How to create a compelling business case for your projects? How to get buy-in and communicate impact? PR is necessary for data teams, even if it’s just internal. Does the rest of the company know who you are? Does your team have a good reputation? Are you trusted advisors? Do you have evangelists who tell your stories for you? In this talk, I will share lessons from my previous career in public relations and marketing and how they apply to your work as a data professional. pyladiescon-2025-78314-public-relations-for-data-teams-how-to-increase-impact-by-building-trust Maggie Wolff en Topics I'll cover: - Why PR matters for Data teams - Balance simplicity and transparency - Tell a story, don't just share numbers - Go from Passive Reporter to Business Driver - Know your audience and how to speak their language - The human behind your work and how to build trust - How to be your own Publicist and get more visibility false https://pretalx.com/pyladiescon-2025/talk/VMQTME/ https://pretalx.com/pyladiescon-2025/talk/VMQTME/feedback/ Main Stream Talk 2025-12-05T21:35:00+00:00 21:35 00:20 We've all been there: staring at a frozen app, a stalled progress bar. Annoying, right? But in medical robotics, a software malfunction can be the difference between success and failure during a surgical study. In **lung cancer diagnoses**, a 3D Convolutional Neural Network (CNN) that is trained to assist is just a promise until a robust MLOps pipeline makes it a reality. A model in a hospital isn't just about accuracy; it's about **unwavering reliability and minimizing downtime** in a secure, on-premise environment. This talk is a deep dive into the code that ensures that trust. pyladiescon-2025-79524-trust-me-i-m-a-robot-mlops-for-medtech Lilinoe Harbottle en This talk is for anyone who has ever seen a brilliant model fail in the real world. We’ll explore a system built with Python’s ecosystem, designed specifically for a critical medical application. You’ll learn how to tackle the most difficult challenges of turning a promising model—like those used in lung cancer diagnoses—into a trustworthy, reliable reality, even in a disconnected environment. Public datasets used for training are often anonymized for patient privacy, which can make it challenging to verify a model's performance across a diverse population. I'll discuss this critical limitation as we explore how to build more robust and inclusive solutions. **What You'll Learn:** * **Predictive Maintenance:** Build a system that detects a component failure before it happens, using telemetry to monitor the "heartbeat" of a surgical robot * **Data Drift Detection:** Ensure your model's accuracy doesn't decay as imaging data changes * **Model Optimization:** Get a resource-heavy 3D CNN to run fast and efficiently, thereby reducing latency and ensuring timely results This talk isn't about theory; it's about **systems development**. Join me to learn what it takes to design, build, and maintain a life-saving system for critical applications. false https://pretalx.com/pyladiescon-2025/talk/YKRHA3/ https://pretalx.com/pyladiescon-2025/talk/YKRHA3/feedback/ Main Stream Talk 2025-12-05T22:00:00+00:00 22:00 00:20 Making faster and memory efficient programs matter. A fast product gives you more customer conversions and shrinks your carbon footprint. To write efficient programs, you need to understand the life cycle of your program inside the server. Where does it live? How does it get executed? What helps it execute faster? Follow the life of your program with Lilliputians - tiny people who can help you understand the flow of your data. pyladiescon-2025-79530-what-you-need-to-know-about-computer-memory-as-a-programmer-as-explained-by-lilliputians Ava Katushka en As a Senior Software Engineer, I have mentored quite a few people. I've noticed that some people lack a clear understanding of how computer memory works, so I created some fun comics that feature essential details of how your program interacts with computer memory. My goal was not to focus on nitty-gritty details, but to get the intuition right. What came out is an illustration so engaging that you can use it to explain the flow of your program to everybody. false https://pretalx.com/pyladiescon-2025/talk/RURANB/ https://pretalx.com/pyladiescon-2025/talk/RURANB/feedback/ Main Stream Talk 2025-12-05T22:40:00+00:00 22:40 00:20 Interactive data exploration tools are excellent for visualizing the data as you are cleaning it, but when a data practitioner analyzes data through drag-and-drop interfaces, the path to reproducibility becomes opaque. This project bridges that gap by capturing UI interactions in the Positron IDE and converting them into clean, readable code across pandas, polars, SQL, and multiple R syntaxes. This talk will include a demonstration of exploring data with a UI and converting that exploration into reproducible code. We’ll walk through the architecture that makes this possible, from tracking UI changes to generating semantically equivalent code across different data manipulation libraries. We’ll also discuss the challenges and considerations that went into the design. This work addresses a critical need in the data science community: tools that enhance usability and reproducibility. Whether you’re building data science tools, analyzing data, or simply frustrated by the gap between exploration and reproduction, this talk will show how thoughtful design can make reproducible science as easy as point-and-click. pyladiescon-2025-79718-making-gui-data-exploration-reproducible-with-python Isabel Zimmerman en false https://pretalx.com/pyladiescon-2025/talk/CCAGCB/ https://pretalx.com/pyladiescon-2025/talk/CCAGCB/feedback/ Main Stream Sponsors's talk 2025-12-05T23:25:00+00:00 23:25 00:15 Enterprise applications need the right data backbone. This session explores two open-source options for powering modern, data-driven applications on Azure: OSS DocumentDB and Azure Database for PostgreSQL. We start with DocumentDB, the open source, MongoDB-compatible engine now governed by the Linux Foundation. You’ll learn how its flexible, JSON-first model accelerates prototyping and iteration, why developers choose it for managing enterprise MongoDB workloads, and how its Azure service adds features such as Autoscale, Entra ID integration, and Customer Managed Keys (CMK). Next, we cover the Azure PostgreSQL database offerings, including Azure Database for PostgreSQL Flexible Server, Elastic Cluster, and the newest enterprise-ready PostgreSQL service. You’ll see how to use these databases support application development, analytics, and AI-driven workloads with reliability and scale. You’ll leave with clear guidance on when to use each database: DocumentDB for flexible, unstructured, and content-centric workloads; PostgreSQL for structured, relational, and analytical ones. Both integrate seamlessly with Python and modern AI frameworks, allowing you to build secure, scalable, and cost-efficient applications without lock-in. Backed by Azure’s managed services, open governance, and unified tooling, these technologies give you the flexibility to choose the right data layer for every enterprise application scenario. pyladiescon-2025-84494-build-enterprise-applications-with-oss-documentdb-and-azure-database-for-postgresql Dingding LuPatty Chow en false https://pretalx.com/pyladiescon-2025/talk/7ZTWMZ/ https://pretalx.com/pyladiescon-2025/talk/7ZTWMZ/feedback/ Main Stream Talk 2025-12-05T23:40:00+00:00 23:40 00:20 In this presentation, my kid and I will talk about building a random route generator to augment the board game Ticket to Ride. Our code chooses cities at random and computes their shortest distance, which translates to points in the game. We explore playful learning and the concepts of computational thinking to learn and implement an algorithm. It’s been a joy to share my love of coding and we are working to improve our project. pyladiescon-2025-79126-ticket-to-ride-python-coding-the-ultimate-expansion-pack-with-a-random-route-generator Fay Shaw en My family loves board games and my 9-year old kid loves beating me. He consistently wins in the game Ticket to Ride. He knows all of the routes in the Europe version so he can block me as soon as I start building routes. In an effort to finally win, I wrote a random route generator in Python. It adds an element of surprise and complexity and we are exploring how we can program together. Together, we are coding the ultimate expansion pack to our favorite board game. Ticket to Ride is a board game where players connect cities with little plastic trains. Players draw destination cards that they need to complete to score points. The point value is determined by the shortest path. Finding the shortest paths for the whole board was a good opportunity to apply the concepts of computational thinking including decomposition, abstraction, and algorithmic thinking. - Decomposition: The entire problem of computing all the shortest paths was too much to start with, so we decomposed it into a smaller problem by looking at a portion of the board. - Abstraction: From there, we abstracted this small map as a graph representation. - Algorithmic Thinking: Together, we used Dijkstra’s algorithm to compute the shortest distances on the small map and then we applied it to our entire map. Finally, we implemented it by writing code in a Jupyter notebook. We found that we really enjoyed playing with random routes and we are working to improve our code through iteration. Even with the modifications, my kid still beats me, so I’m just going to have to play better. The best part of all this has been sharing my love of coding in a playful way. false https://pretalx.com/pyladiescon-2025/talk/C7W8QM/ https://pretalx.com/pyladiescon-2025/talk/C7W8QM/feedback/ Main Stream Talk 2025-12-06T00:40:00+00:00 00:40 00:20 Whether your working in data, web development, automation or embedded systems, you've probably had a time when you felt like it was team challenges, not technical challenges, holding you back. Hopefully you've also had a time when you had a great team solving tough problems, where the team made everything feel easier. In this talk we'll look at how an understanding of group development models can help improve your technical team experiences - even if you're not the leader of the team. pyladiescon-2025-79414-strengthen-technical-teams-through-group-development-models Lydia Peabody en In this talk we'll focus on Tuckman's model of group development, which argues that groups go through five stages in a cyclic fashion: forming -> storming -> norming -> performing -> adjourning. In each of these stages the dynamics and needs of the team are different. Successful teams move through these stages, and spend most of their time in "performaing" - working well together to accompish goals. Teams that feel tough are often stuck in storming, where it's unclear who's responsible for what and thing's don't "click". If you can learn to identify what stage your team is at, you can help nudge it in the right direction to move towards performing. As a team leader you can have significant influence on these transitions, but even as team members our actions and input can help steer the whole team in useful directions. We'll work through common scenarios you might experience (for example, a brand new team starting up, a new developer joining an existing team, a change in manager, team member leaving), look at what's happening in the team in each case. We'll share strategies for how both team leaders and team members can use the group development model to help the group work through change and get back to being a fun, effective team as quickly as possible. false https://pretalx.com/pyladiescon-2025/talk/HNZPGJ/ https://pretalx.com/pyladiescon-2025/talk/HNZPGJ/feedback/ Main Stream Sponsors's talk 2025-12-06T01:05:00+00:00 01:05 00:15 This talk demystifies Large Language Models (LLMs) by presenting them not as magic, but as powerful APIs that developers can master using existing software engineering skills. It advocates for an engineering-based approach, demonstrating how to use Python's abstract base classes to create a vendor-agnostic and testable interface. This allows for interchangeable backends, from cloud services like Gemini to local mocks, making AI development more modular and robust. The talk empowers developers to look past the hype, and instead apply their existing knowledge to build reliable and innovative applications, concluding with the message that with LLMs, "It's just an API. It's just Python. Grab a key, write a test, and build the future." pyladiescon-2025-85066-beyond-the-hype-llms-aren-t-magic Anna Eilering en his talk by Anna Eilering, a Developer Relations Engineer at Google's Open Source Programs Office, cuts through the hype surrounding Large Language Models (LLMs) and Generative AI. The core message is that LLMs are not magic, but powerful APIs that developers can and should work with just like any other code. The talk is aimed at Python developers who may feel intimidated by the mystique around AI, and encourages them to leverage their existing skills to build innovative solutions. false https://pretalx.com/pyladiescon-2025/talk/9HN9JG/ https://pretalx.com/pyladiescon-2025/talk/9HN9JG/feedback/ Main Stream Keynote 2025-12-06T01:20:00+00:00 01:20 00:30 The free and open-source software community is an astonishing global feat of human collaboration, independently of the software that it creates. Taking a more active role in this community offers significant potential benefits, but also plenty of opportunities for frustration and disappointment. Sparing some time for personal reflection on the question posed in this talk can go a long way towards making our community interactions a positive experience for all involved. pyladiescon-2025-82562-do-you-want-to-get-more-involved-in-open-source Alyssa Coghlan en false https://pretalx.com/pyladiescon-2025/talk/SXJPB7/ https://pretalx.com/pyladiescon-2025/talk/SXJPB7/feedback/ Main Stream Talk 2025-12-06T01:50:00+00:00 01:50 00:20 Learn how to build open-source bilingual dictionaries for Indonesia’s local languages using Python NLP libraries such as spaCy, NLTK, and Transformers. This talk explores practical methods to process low-resource languages, including hybrid rule-based and transformer models. We will discuss how to build dictionary data pipelines and add features like search, synonyms, antonyms, and lemmatization. The session also highlights the role of open-source collaboration in preserving linguistic diversity through technology. pyladiescon-2025-79221-building-digital-dictionaries-for-local-languages-in-indonesia-with-python-nlp Muh Naufal Muzhaffar en Discover practical techniques for building bilingual digital dictionaries (Indonesian ↔ local languages) using Python NLP libraries such as spaCy, NLTK, and Transformers. This talk will cover how to design and implement hybrid models combining rule-based and machine learning approaches, build efficient data pipelines for dictionary management, and integrate advanced linguistic features like synonym and antonym lookup, lemmatization, and full-text search. Additionally, it emphasizes fostering sustainable, community-driven projects that empower local language preservation and digital accessibility. false https://pretalx.com/pyladiescon-2025/talk/DQB3PR/ https://pretalx.com/pyladiescon-2025/talk/DQB3PR/feedback/ Main Stream Keynote 2025-12-06T02:55:00+00:00 02:55 00:30 **The tech is the easy part; people are the real challenge. Machines follow instructions. People follow principles — and that’s what makes leadership both hard and meaningful. Communities rarely fall apart because people lack passion; they fall apart because they lack systems and resources.** When Matt and I first attended PyCon PH 2012, we didn’t plan to lead anything. We were just two young developers who found a community that felt alive. In 2013, with seven others, we incorporated PythonPH to keep that spark going. What followed were years of learning the hard way: burnout, stretched volunteers, and the truth that passion alone can’t sustain a community. Early on, it takes unsustainable effort before anything feels stable. As co-founders stepped back — the way people naturally do when life and priorities shift — we stayed because someone had to. But when responsibility quietly piles onto the same few people, it wears them down. That’s how burnout really happens. Over time, we learned that communities flourish not because of exceptional leaders, but because of systems that make leadership shareable. When the structure makes it easy for committed people to step up, sustainability finally follows. This talk is for anyone who carries the invisible load of keeping a community alive — especially those who quietly hold things together. The emotional labor, the coordination, the reliability. These things are rarely seen, but always felt. And showing up matters; you never know who’s drawing strength from your example. For many women, that quiet kind of role modeling will feel familiar. If you’re stretched thin but still hopeful, this talk is for you — a reminder to pause, reconnect with what matters, and lead from your values. The work is tough, but rewarding, and you don’t have to carry it alone. pyladiescon-2025-82833-scaling-and-systematizing-grassroots-leadership-lessons-from-pythonph-and-pyladies-manila Micaela Reyes en Grassroots leadership often begins with passion, but passion alone can’t keep a community alive. After more than a decade of building PythonPH and PyLadies Manila, we learned that communities don’t flourish because of exceptional leaders, they flourish because of systems that make leadership shareable. When you create structures that make it easier for committed people to step up, sustainability finally begins to take shape. This talk is a practical and honest look at what it really takes to build a volunteer-driven community that can outlast its founders. We’ll talk about the early years of unsustainable effort, the realities of burnout, and the turning points that forced us to rethink how leadership works. We’ll also explore the Kaizend approach: a culture and values-driven model for supporting, developing, and retaining volunteers. Part of that work meant accepting that even nonprofits need business-level clarity — resources, budgeting, and operations — to stay healthy and fair to volunteers. If you’ve ever felt stretched thin but still hopeful, this session is for you. I’m just sharing what we learned the hard way, and hopefully it sparks ideas you can shape into something that fits your own community. It’s a reminder to pause, reconnect with what matters, and lead in a way that honors your values. The work is tough, but fulfilling, and you don’t have to carry the load alone. **1. Systems make leadership sustainable.** I learned the hard way that communities flourish when leadership is shared, not concentrated. Simple systems made it easier for committed people to step up, and that’s when things slowly started to become sustainable. **2. Burnout is a design problem, not a personal failure.** Early on, it takes unsustainable effort before anything feels stable, mostly because you’re learning faster than you’re structuring. But once we started spreading responsibility and building systems around the work, things slowly got lighter. Burnout wasn’t a sign that we were weak; it was a sign that the system needed to change. **3. Care doesn’t have to depend on vibe. It can be intentional and structured.** Through Kaizend, I realized that volunteer care can’t be left to chance. The Salesian Oratory model (Home, School, Spirituality, Playground) helped us build a system that supports the whole person, and that foundation became our real source of resilience. It’s what inspired us to build Kaizend. **4. Culture and values are your real levers.** I learned that what works for one community won’t always work for another. Filipino Bayanihan, our cultural instincts, and our shared principles shaped the systems that ended up working for us. One-size-fits-all models rarely work. Leading with your community’s values is what helps you build something that lasts. **5. Even movements and nonprofits need to be run like a business.** I learned that even nonprofits need clarity in budgeting, planning, and operations. Passion drives the mission, but structure keeps the work fair and sustainable for volunteers. false https://pretalx.com/pyladiescon-2025/talk/QDUYTD/ https://pretalx.com/pyladiescon-2025/talk/QDUYTD/feedback/ Main Stream Panel 2025-12-06T03:50:00+00:00 03:50 00:30 Across Asia, many PyLadies chapters are building vibrant communities where women can grow in tech, connect with peers, and develop new skills. At the same time, chapters often face shared challenges, such as securing funding, increasing visibility, and sustaining long-term engagement. In this panel, we bring together PyLadies leaders from across Asia to share our experiences, celebrate successes, and discuss the hurdles we still face. We will explore practical strategies to strengthen collaboration, support one another across borders, and amplify the impact of our chapters. This conversation is an opportunity for all of us to learn from one another, spark new ideas, and build stronger connections that help grow a more united and thriving PyLadies network across the region. pyladiescon-2025-84087-python-across-asia-collaboration-and-cross-border-support Silvia Larasatul MDaphne ChoongEllaine Dela CruzCathy HuiJashlein Leanne Marquez en false https://pretalx.com/pyladiescon-2025/talk/RLU8QR/ https://pretalx.com/pyladiescon-2025/talk/RLU8QR/feedback/ Main Stream Talk 2025-12-06T04:20:00+00:00 04:20 00:20 Governments generate vast amounts of data, but making it accessible, actionable, and engaging for the public remains a challenge. This session explores how the City of Austin, Texas, USA leverages Python to transform open data into interactive dashboards that foster transparency and empower citizens. Through real-world examples from Austin’s public sector projects, attendees will learn how data visualization and analytics can strengthen community engagement, enhance decision-making, and create a more informed society. pyladiescon-2025-80339-democratizing-data-python-powered-dashboards-and-open-data-for-transparent-governance-in-austin Tanvi Sharma en The City of Austin, Texas, USA is committed to promoting transparency and collaboration by using data to empower its residents. As the Open Data Portal Liaison for Austin's Parks and Recreation Department, I have firsthand experience in leveraging Python to turn city data into powerful tools for public engagement. This talk will showcase: 1. Using Python to Access and Process Open Data - Introduction to the City of Austin’s Open Data Portal. - Demonstrating how Python libraries like Pandas, GeoPandas, and NumPy help in cleaning and transforming raw datasets for analysis. 2. Building Interactive Dashboards - Real-world examples of dashboards built for utility management, park planning, and public project monitoring. - Tools like Plotly Dash and Streamlit for creating engaging visualizations that communicate complex data to diverse audiences. 3. Enhancing Transparency Through Data - How Python-powered analytics promote better governance by making city metrics—like resource allocation and project timelines—available to all stakeholders. - Insights from Austin’s Parks and Recreation Department projects, including the creation of dashboards for asset management and park accessibility, which improved both operational efficiency and public trust. 4. Live Demonstration - A step-by-step walkthrough of transforming an open dataset (e.g., park usage or community feedback) into an interactive dashboard. - Tips and tricks for making dashboards accessible and visually appealing to non-technical audiences. 5. Call to Action - Encouraging Python developers to engage with open data in their communities. - Highlighting the importance of open-source tools in fostering equitable access to data. Whether you are a Python developer, a public sector professional, or simply curious about how technology can bridge the gap between governments and communities, this talk will provide actionable insights and practical skills to make data work for the people. Join me to learn how Python is reshaping the future of governance in Austin, Texas, USA and beyond. false https://pretalx.com/pyladiescon-2025/talk/JXYZEN/ https://pretalx.com/pyladiescon-2025/talk/JXYZEN/feedback/ Main Stream Talk 2025-12-06T04:45:00+00:00 04:45 00:20 Forecasting isn’t just about numbers, it’s about making the future understandable and actionable. TimeCopilot (https://timecopilot.dev/) is an open-source Python agent that combines large language models with time-series foundation models to automate, explain, and interact with forecasts. In this talk, I’ll share how we orchestrate Python-based agent workflows with tools like PydanticAI and Hugging Face to bridge the gap between statistical accuracy and human-friendly insights. We’ll walk through real examples, lessons learned from building for the open-source community, and how agentic systems can make forecasting more accessible to everyone. pyladiescon-2025-79795-timecopilot-the-genai-forecasting-agent Azul Garza en TimeCopilot is built in Python to bring together two powerful worlds: the reasoning capabilities of LLMs and the precision of time-series foundation models like AWS Chronos, Google TimesFM, Salesforce Moirai, and Nixtla TimeGPT. It’s model-agnostic, supports ensembles for more robust results, and lets users ask questions about their data in plain English. This session will cover: - How Python tools (PydanticAI, PyTorch, Hugging Face) enable agentic forecasting workflows. - Designing prompts and orchestration pipelines that integrate LLM reasoning with domain-specific models. - Making forecasts explainable and interactive for technical and non-technical audiences. - Lessons from our open-source journey, including our 1st-place on the Salesforce GIFT-Eval benchmark. Whether you’re into data science, AI agents, or building inclusive tools, you’ll leave with ideas and code you can use to create your own Python-powered forecasting agents. false https://pretalx.com/pyladiescon-2025/talk/QQRHRF/ https://pretalx.com/pyladiescon-2025/talk/QQRHRF/feedback/ Main Stream Talk 2025-12-06T05:05:00+00:00 05:05 00:20 From Volunteer to Organizer: Start Small, Dream Big I began my journey with PyLadies Seoul as a volunteer at PyCon KR, and quickly resonated with the community’s mission of building an inclusive space for women developers. Inspired, I joined as an organizer, contributing to our year-end celebration, leading the Fluent Python study group, and participating in PyCon KR 2025. Along the way, I connected with PyLadies chapters from other countries and collaborated with fellow organizers to drive meaningful change. Together, we successfully rebooted PyLadies Seoul, creating a vibrant, welcoming space for women from diverse backgrounds. This talk shows how small steps of engagement can grow into leadership and impact. If you’ve ever wondered how to move from participant to contributor, my story demonstrates that starting where you are—with curiosity and care—can make a difference. pyladiescon-2025-80319-from-volunteer-to-organizer-start-small-dream-big Hwayoung Cha en false https://pretalx.com/pyladiescon-2025/talk/MU3PDT/ https://pretalx.com/pyladiescon-2025/talk/MU3PDT/feedback/ Main Stream Talk 2025-12-06T05:25:00+00:00 05:25 00:20 When beginners start learning Python, they often understand the idea of fixing errors but don’t fully connect it to the hands-on process of debugging. I learned this firsthand while building a production application, where the project’s size and complexity pushed my debugging skills to the limit. In this talk, I will break down a beginner-friendly toolkit for finding and fixing bugs in Python. We’ll start with making sense of error messages and learning how to read a traceback without panicking. Attendees will explore how to use tools like print() and assert strategically, and why exceptions and logging can be lifesavers when things go wrong. We will also tackle common beginner pitfalls, from harmless typos to sneaky logical errors, and share practical techniques for spotting them faster. By the end, attendees will walk away not just knowing how to debug, but with the confidence to face errors head-on, whether working on a small script or a large project. pyladiescon-2025-79620-so-it-broke-now-what-debugging-python-101 Rahmat Akintola en By using examples from my journey, including a production app where debugging became my daily workout, this talk aims to help attendees understand debugging and develop the skills to approach errors with confidence. ## Breakdown of session: - *Introduction* - 2 mins Quick story on how I learned debugging the hard way, why it’s a core dev skill, and what we’ll cover. - *What Debugging Means* - 2 mins More than “fixing errors” — debugging is a process of understanding what’s happening when code breaks, why it happened, and how to prevent it. - *Using Print Statements* - 2 mins How to use print() statements and how too much of it can make the codebase messy. - *Raising Exceptions* - 2 mins Writing programs to anticipate and recover from errors using raise to signal problems and try/except to handle them gracefully. It shows how this makes debugging faster and programs more robust. - *Logging* - 2 mins Why logging is important in larger projects. Quick setup with Python’s `logging` module, choosing log levels, and using logs to trace bugs after deployment. - *Assertions* - 2 mins Spotting problems early with assert as a sanity check to make sure your code isn’t doing something wrong - *Using a Debugger* - 3 mins A peek at Python built-in debugger [pdb](https://docs.python.org/3/library/pdb.html) and IDE debuggers to step through code and see what’s happening. - 1 min – Wrap-up Recap of debugging techniques, encouragement to practice regularly, and a mindset shift from “errors are scary” to “errors are clues.”. false https://pretalx.com/pyladiescon-2025/talk/YL3LZC/ https://pretalx.com/pyladiescon-2025/talk/YL3LZC/feedback/ Main Stream Talk 2025-12-06T05:50:00+00:00 05:50 00:20 When we define our own classes in Python, they often don’t behave like built-in types especially when we try to use operators like +. In this short talk, we’ll look at how to fix that by using Python’s special methods. With a hands-on example using \_\_add\_\_ , we’ll see how to make our objects work the way we expect, and why this improves code readability and maintainability. A useful tool in your Python toolkit, even if you're not writing mathematical code. pyladiescon-2025-79458-can-we-add-objects-the-way-we-do-for-numbers Swarooprani Manoor en Custom Python classes don’t behave like built-in types by default. For example, if you try using + on your own objects (obj1 + obj2), it won’t work unless you tell Python what to do. In this short and beginner-friendly talk, we’ll see how to make operators like + work for user-defined classes using special methods like \_\_add\_\_. With a simple example, you’ll learn how this improves code readability, makes your classes easier to use, and helps your code feel more Pythonic. false https://pretalx.com/pyladiescon-2025/talk/HBMEYA/ https://pretalx.com/pyladiescon-2025/talk/HBMEYA/feedback/ Main Stream Talk 2025-12-06T06:10:00+00:00 06:10 00:20 Around the world, billions of dollars are flowing into nature-based solutions to tackle climate change, biodiversity loss, and land degradation. But ensuring that these investments deliver real, measurable results is a major challenge. At the World Resources Institute (WRI), our Restoration Monitoring team is using Python to transform how restoration progress is tracked, verified, and shared, with the goal of bringing greater trust and transparency to the movement. This talk will showcase how we combine Python’s rich ecosystem of data science, geospatial, and AI/ML libraries to power the monitoring, reporting, and verification (MRV) of restoration activities under WRI’s TerraFund program. Since 2022, TerraFund has channeled $32 million to nearly 200 restoration champions. I’ll walk through our rule-based decision support system—a Python tool built with requests and API-specific client libraries—that integrates multiple data streams to determine which projects should undergo remote or field verification. This approach helps us reduce uncertainty, optimize field missions, and improve cost planning. As part of our remote verification protocol, we deploy state-of-the-art computer vision models on high-resolution Maxar imagery to label trees and saplings with bounding boxes. Our data production process uses Python-based pipelines to query, order, and process imagery, then run inference to detect trees. This enables automated tree counting at scale, turning terabytes of imagery into actionable project-level verification. Finally, results are integrated into TerraMatch, our end-to-end platform connecting funders and restoration champions, and making insights about progress publicly accessible. Attendees will learn how Python acts as the connective tissue across APIs, AI/ML models, and geospatial workflows, and how these tools are helping us build scalable, transparent monitoring systems. pyladiescon-2025-79552-building-trust-in-the-restoration-movement-with-python-ai-and-satellite-data Jessica Ertel en false https://pretalx.com/pyladiescon-2025/talk/JV7TG9/ https://pretalx.com/pyladiescon-2025/talk/JV7TG9/feedback/ Main Stream Special 2025-12-06T06:30:00+00:00 06:30 00:30 The holiday season can be overwhelming, filled with obligations and competing demands. This session offers a short, values‑based exercise to help participants reconnect with what matters most. Using guided reflection on joy, pride, and fulfillment, attendees will identify their core values and translate them into a simple statement and action plan. By grounding in values, participants can clarify focus, set boundaries, and navigate the season with greater steadiness and meaning. Led by Dr. Drew B. Mallory, professional coach and professor of management and organizational behavior and executive coach, this session draws on his expertise in intersectional inclusion, wellness, and executive coaching. Dr. Drew specializes in helping individuals and organizations align actions with values, making this exercise both practical and deeply relevant for navigating stressful times. pyladiescon-2025-84164-from-stress-to-strategy-rediscovering-your-values-with-coach-dr-drew-b-mallory Dr Drew B. Mallory en The holidays often bring stress, distraction, and pressure to meet others’ expectations. This session, which I use with both students and coachees, provides a structured, values‑based method to pull yourself out of overwhelm and determine where to focus. Drawing on research that shows values predict behaviors, participants will: -Reflect on moments of joy, pride, and fulfillment to uncover personal values. -Select 3–5 values that feel most relevant for this season. -Write a short value statement that can guide decisions and boundaries. -Translate values into micro‑actions and rituals that restore clarity and connection. By the end, you will leave with a personalized values statement and a one‑page plan for applying it during the holiday season. This exercise is practical, brief, and adaptable, helping you navigate stress by acting from your values rather than reacting to circumstances. About the facilitator: Dr. Drew B. Mallory is a private professional coach and professor at Sasin School of Management in Bangkok and a global consultant on inclusion and organizational behavior. His work bridges scholarship and practice, empowering people to act from their values in both personal and professional contexts. With experience coaching executives and designing tools for cross‑cultural competence, Dr. Drew brings a pragmatic, research‑based approach to this exercise—ensuring participants leave with strategies they can apply immediately during the holiday season. Learn more about this and other self-development methods at www.coachwithdrew.today false PyLadiesCon Rediscovering Your Values 2025 https://pretalx.com/pyladiescon-2025/talk/DLFAL8/ https://pretalx.com/pyladiescon-2025/talk/DLFAL8/feedback/ Main Stream Special 2025-12-06T07:20:00+00:00 07:20 00:30 Weather you are starting your day or needing to rewind and unwind from the chair that you glued to by all the captivating talks.. this will feel great for you! We will do a short functional yoga flow to help reset and charge up for the rest of the conference. A mat is welcome, but you can also do this with a chair. No special clothes or other equipment needed. pyladiescon-2025-84165-feel-good-yoga Tereza Iofciu en false https://pretalx.com/pyladiescon-2025/talk/TV8UVR/ https://pretalx.com/pyladiescon-2025/talk/TV8UVR/feedback/ Main Stream Talk 2025-12-06T07:50:00+00:00 07:50 00:20 In this talk, I will introduce a prototype that allows a small car powered by a Raspberry Pi Pico to be controlled using multimodal input such as voice and images. Voice commands and hand-drawn routes are entered via Gradio and converted into instructions like “move forward” or “turn right” using LangChain. The Pico acts as an HTTP server, receives these commands over Wi-Fi, and controls the car accordingly. I’ll share the system structure and technologies used, along with the fun and creative challenges I encountered during development. It’s a fun project that combines generative AI and IoT! --- 本発表では、音声や画像といったマルチモーダルな入力から、Raspberry Pi Picoで動く小型カーを制御する試みを紹介します。 構成としては、Gradioで音声や手描きルートを入力し、それをLangChainを使って「前進」「右折」などを意味するコマンドに変換。Picoは簡易的なHTTPサーバとして動作し、Wi-Fi経由で指示を受けてカーを制御します。 本発表では、この構成と技術要素を共有しながら、実際に試してみて感じた楽しさや工夫した点をお伝えできればと思っています。 ゆるく楽しく「生成AI × IoT」の可能性を探る取り組みです！ pyladiescon-2025-78573-go-straight-then-turn-right-how-i-built-a-voice-controlled-toy-car-using-generative-ai-gradio-and-raspberry-pi-pico Tomoko Furuki（@komo_fr） ja Wouldn't it be fun to control a small car using voice commands or hand-drawn routes? This idea led me to try a playful experiment combining generative AI and IoT. In this prototype, I used the following tools and technologies: - A commercially available Pico car kit with a **Raspberry Pi Pico WH** - **Gradio** for voice input and route drawing - **LangChain** to convert user instructions like “Go straight” or “Turn right” into commands and send them to the car - An HTTP server using **Microdot**, running on the Pico to receive commands via Wi-Fi In this talk, I’ll walk through the technical flow while also highlighting some fun aspects unique to generative AI—such as how the car responds to vague or playful commands like “Dance!” or “You’re so cute!” Even without deep technical knowledge, this is a fun prototyping example you can try using Python and familiar tools. **Note: This talk will be presented in Japanese, with slides in both English and Japanese.** --- 音声指示や手描きルートを使って小型カーを操作できたら面白そう！ そんな発想から始まった、生成AIとIoTを組み合わせたゆるい実験を紹介します。 今回のプロトタイプでは、以下のようなツールや技術を使っています。 - 車体は市販のPicoカーキット + **Raspberry Pi Pico WH** を使用 - **Gradio** で音声入力やルート画像を入力 - **LangChain** でユーザの指示を「前進」「右折」などを意味するコマンドに変換し、Picoカーに送信 - Pico上で **Microdot** を使ってHTTPサーバを動かし、Wi-Fi経由でコマンドを受信 本発表では、こうした技術的な流れを紹介しつつ、「踊ってみて！」「かわいいね」といった曖昧な言葉にも反応する、生成AIならではの面白さもあわせてお話しします。 難しい知識がなくても試せる、Pythonと身近なツールを使ったプロトタイピングの楽しさを共有します。 **注意：この発表は日本語で行い、スライドは英語と日本語を併記します。** false https://pretalx.com/pyladiescon-2025/talk/33W8ES/ https://pretalx.com/pyladiescon-2025/talk/33W8ES/feedback/ Main Stream Talk 2025-12-06T08:10:00+00:00 08:10 00:20 What if the tools we use to predict disease risk don't work for entire populations, simply because very few people (studies) test them over there? That's the case for many polygenic risk score (PRS) models and pipelines, which often fail when applied to African genomes. PRS is a number that estimates how likely someone is to develop a disease based on their DNA. In this talk, I'll walk through how I am using Python to build something better: a reproducible pipeline for genomic tools that center inclusion. More than a technical talk, this is a story about learning, persistence and impact. I'll share how I use accessible Python libraries like argparse, pandas, subprocess and shutil to automate preprocessing, handle containerised genomic tools with Docker, and stitch everything together using workflows like Nextflow. The tools I'm building are still a work in progress, but this talk will spotlight the imperfect, iterative process of building (while learning) pipelines that don't exclude underrepresented populations by design. I'll share how Python makes it possible to go from fragmented data to actionable results. This talk is for anyone curious about bioinformatics, passionate about global inclusion or simply learning Python pyladiescon-2025-78784-python-in-africa-s-dna-building-inclusive-and-reproducible-genomic-tools Chioma Oselu (Onyido) en This talk is a personal and technical journey into using Python as a tool for representation in genomics. It shares the story of how I (under the guidance of excellent mentors) began building a pipeline that challenges the norms of who gets represented in disease risk prediction tools. Many widely used tools (in genomics) for analysing DNA, like those used to calculate polygenic risk scores (PRS), perform poorly on African datasets. In addition to data scarcity, this is a design issue rooted in various assumptions, some of which were never questioned. This talk highlights how I'm using Python to answer that question, as a learner using accessible tools, libraries and open-source principles to create change from my corner of the world. 5 KEY POINTS THIS TALK WILL COVER - Relevance of Talk: I begin by grounding the audience in the problem: how underrepresentation of African genomes leads to poor health predictions and inequitable science. Most people outside of genetics don't realise that common tools perform dramatically worse on diverse populations, and this can have serious consequences for things like disease risk scoring, diagnosis and treatment decisions. The problem extends beyond data diversity, it lies in the inflexibility of existing pipelines to process such diversity effectively. Many genomic tools weren't built to work for African data, and in some cases, they completely breakdown. I'll share a few brief, easy-to-understand examples of where this happens in real datasets. - How Python Helps Me Build Solutions: The heart of the talk is a practical, beginner-friendly look at how Python enables me to solve this problem, step-by-step. I'll introduce the libraries I use daily: i. argparse: For making my scripts flexible and user-friendly ii. pandas: For cleaning, transforming and exploring genetic summary statistics and phenotype data iii. subprocess: For calling Docker containers and orchestrating commands programmatically iv. Shutil: For managing folders and moving data during preprocessing Finally, a quick mention of how this integrates with workflow tools like Nextflow - Learning in Public, Building While Learning: One of the key themes of this talk is that, you don't have to be an expert to build something impactful. I'll share: - What it looked like to go from having basic Python knowledge to writing scripts that trigger enter bioinformatics pipelines - The many (many!) mistakes I made while figuring out how to manage data from diverse sources - How I broke the big problem down into scripts and containers, even while learning pipeline management, Docker and bash along the way. - Why starting before you're ready is the most powerful thing you can do as a Python learner. - Challenges I'm Solving With Python: Instead of detailing the benchmarking methodology, I'm spotlighting the problems my code is built to solve: - Handling messy metadata: How python scripts help me detect and resolve inconsistent column names and formats across datasets. - Reproducibility on a shoestring: Why using Python + Docker gives me the ability to share reproducible, portable pipelines. - Debugging inclusivity: How scripting helped me spot patterns of exclusion and fix issues that most pipelines would miss (e.g. default parameters that don't suit African LD structures or allele frequency ranges). All of this is still a work in progress, and that's part of the point. - Significance of Talk: We often celebrate tools and technologies after they've become polished and complete. But in this talk, I want to celebrate the act of building while learning, especially when it's in service of equity and inclusion. For too long, people from underrepresented regions and learners new to coding have been made to feel like they don't belong in technical spaces unless they're already experts. This talk says otherwise. Python isn't just a programming language. For me, it's a way to claim space, ask hard questions, and build tools that make health and data inclusive for all. false https://pretalx.com/pyladiescon-2025/talk/3KNKHS/ https://pretalx.com/pyladiescon-2025/talk/3KNKHS/feedback/ Main Stream Talk 2025-12-06T08:35:00+00:00 08:35 00:20 Once upon a time, if you wanted to build a modern web experience using a Python based web framework, an obvious choice was to expose a bunch of APIs and build your frontend in something else entirely (eg React). Times have changed. Tools like HTMX and AlpineJS have made HTML great again. This talk is a demonstration and explanation of the power of these tools. pyladiescon-2025-79495-modern-frontend-development-the-simple-way Sheena en Once upon a time, if you wanted to build a modern web experience using a Python based web framework, an obvious choice was to expose a bunch of APIs and build your frontend in something else entirely (eg React). Times have changed. Tools like HTMX and AlpineJS have made HTML great again. These tools make full stack web dev fast and fun, they simplify development, shrink code bases, and give backend devs the ability to do modern frontend work. In my personal experience I've seen the adoption of these tools lead to massive productivity gains for junior and senior devs alike. false https://pretalx.com/pyladiescon-2025/talk/89CQX7/ https://pretalx.com/pyladiescon-2025/talk/89CQX7/feedback/ Main Stream Sponsors's talk 2025-12-06T08:55:00+00:00 08:55 00:15 Most of us have had that moment where we set up a machine or a dev environment “just this once,” promise ourselves we’ll document it later, and then… never do. Now imagine doing that across thousands of servers, multiple data centres, and a bunch of very latency-sensitive systems. Terrifying, right? In this talk, I’ll share how we tackled this problem at Optiver by building our own Infrastructure as Code platform — for physical, bare-metal machines that absolutely refuse to behave like the cloud. Along the way, we learned a lot about declarative design, reconciliation loops, standards, Python tooling, and also, human nature. You don’t need to work in trading or infrastructure (or know what a picosecond is!) to follow along. If you’re curious about building systems you can trust, reducing manual chaos, or just want to hear an honest story of how code can help tame messy realities, I hope you’ll take something valuable away. pyladiescon-2025-85045-infrastructure-as-code-in-a-latency-critical-bare-metal-world Leanne Fok en false https://pretalx.com/pyladiescon-2025/talk/7XR3TZ/ https://pretalx.com/pyladiescon-2025/talk/7XR3TZ/feedback/ Main Stream Talk 2025-12-06T09:10:00+00:00 09:10 00:20 Open source runs on passion, but passion is not a renewable resource. This talk will explore the hidden emotional and social costs of contributing to open-source projects. From burnout to invisibility, we’ll reflect on how the very communities we depend on can unintentionally exhaust the people keeping them alive. Attendees will walk away with practical ways to support contributors, make communities more welcoming, and build open source that truly includes everyone. pyladiescon-2025-79363-not-just-code-building-communities-that-don-t-burn-people-out Aishat Muibudeen en Open source has revolutionized how we build software; however, it hasn’t solved the problem of how we support the people who make it possible. Behind every feature request, GitHub issue, or Slack ping is a human being trying to balance their desire to help with their mental, emotional, and sometimes financial limits. And while we celebrate open source wins, we rarely discuss the cost of participation: burnout, loneliness, exclusion, and the pressure to be “always on.” This talk takes a critical yet hopeful look at the human realities of contributing to and maintaining open-source projects. We’ll explore: * The hidden labor behind community work and code maintenance * How issues like burnout, gatekeeping, and lack of diversity still persist * Why community care should be treated as infrastructure * What inclusion actually looks like (beyond just opening a repo) * Practices that support contributor well-being without sacrificing project momentum * Whether you're a project lead, data scientist, contributor, or community manager, this session will help you see the people behind the pull requests and understand how to build spaces where they can thrive, not just survive. false https://pretalx.com/pyladiescon-2025/talk/7UL7TG/ https://pretalx.com/pyladiescon-2025/talk/7UL7TG/feedback/ Main Stream Panel 2025-12-06T09:40:00+00:00 09:40 00:30 Mentorship is at the heart of many tech journeys, often being the first place where women feel supported, guided, and encouraged to grow. In this panel, we will hear stories from across Africa, showing how women start as mentees and eventually step into mentorship themselves, carrying forward the support they once received. Panelists will share experiences from different types of mentorship, including structured programs, peer-to-peer learning, and informal guidance, and how these relationships help develop technical skills as well as leadership, advocacy, and community-building abilities. The discussion will also highlight how mentorship is a two-way street, with mentees shaping the growth and perspective of their mentors. Finally, we will explore practical advice for new mentees, strategies for mentors to provide meaningful support, and ideas for expanding mentorship programs to reach and empower even more women in tech. pyladiescon-2025-84088-mentorship-that-works Aisha BelloCecilia TivirSumaiya NalukwagoDorothy Ewuah en Sumaiya Nalukwago is an experienced IT Professional and Tech Community Lead with over six years of proven success in leveraging technology to drive economic development and foster innovation across Uganda. A recognized voice in the continental tech landscape, Sumaiya has been a dedicated volunteer for PyCon Uganda since its inception and has spoken at major international conferences. This experience underscores her deep commitment to the open-source movement and her ability to engage diverse global audiences. As a Community Lead for Google Developer Groups Cloud Mbarara and an Ambassador for Google Women Techmakers, Sumaiya actively skills and mentors thousands of youth and women entrepreneurs in digital literacy, business development, and tech-solution design. She specializes in translating complex technological concepts into accessible, actionable skills, enabling grassroots enterprises to thrive in the digital economy and confidently bridge the digital skills gap. false https://pretalx.com/pyladiescon-2025/talk/UELPHB/ https://pretalx.com/pyladiescon-2025/talk/UELPHB/feedback/ Main Stream Talk 2025-12-06T10:10:00+00:00 10:10 00:20 Reasoning is one of the most powerful—and sometimes surprising—capabilities in large language models. In this talk, we’ll demystify what “reasoning” means, how intermediate steps improve problem-solving, and why certain techniques—like Chain-of-Thought, Self-Consistency, RL-based reasoning, and distilled reasoning models—can unlock smarter behavior. We’ll also explore how reasoning can emerge naturally through reinforcement learning. Through practical examples in generative and agentic AI, we’ll cover when reasoning delivers value, its trade-offs, and how to optimize context for better performance. Attendees will leave with a clear framework for applying reasoning models effectively. pyladiescon-2025-79439-from-chain-of-thought-to-agentic-ai-demystifying-reasoning-in-llms Irene Donato en Large language models (LLMs) have advanced far beyond text prediction—they now exhibit surprising abilities to reason through problems, follow multi-step processes, and produce logically consistent answers. Yet “reasoning” in AI often means something different from human reasoning, and understanding that distinction is key to using it effectively. This talk clarifies what reasoning means for LLMs and why intermediate steps matter for complex problem-solving. We’ll explore Chain-of-Thought prompting, Self-Consistency decoding, RL-based reasoning, and distilled reasoning models—highlighting their strengths, trade-offs, and performance costs. A particular focus will be on adaptive reasoning models, which dynamically adjust their computational depth based on problem complexity. Unlike traditional fixed-computation approaches, these models can allocate more "thinking time" to harder problems while remaining efficient on simpler tasks. We'll then present a practical decision framework for when to apply reasoning—balancing value, complexity, and latency. Not all tasks benefit from multi-step inference; some work best with direct outputs, while others gain from deeper reasoning. We’ll also address challenges: compound errors cascading through steps, trade-offs between model size and reasoning quality, and scaling issues with large context windows. Finally, we’ll cover context engineering—structuring prompts, retrieval pipelines, and context windows to improve reasoning reliability and efficiency. Attendees will leave with: - A clear understanding of reasoning in LLMs. - Insight into reasoning emerging from RL with verifiable output. - Knowledge of key techniques and emerging architectures. - Practical tips for optimizing reasoning in AI projects. false https://pretalx.com/pyladiescon-2025/talk/ZGPVCU/ https://pretalx.com/pyladiescon-2025/talk/ZGPVCU/feedback/ Main Stream Talk 2025-12-06T10:30:00+00:00 10:30 00:20 In Africa, getting inclusive education is still quite difficult, especially for kids with learning difficulties like dyslexia. Imprecise word comprehension and generally subpar reading performance are hallmarks of dyslexia, a neurological disease. It impacts a sizable portion of school-age children, with a higher prevalence in boys, putting them at lifetime risk for subpar academic achievement and low self-esteem. The United Nations Educational, Scientific, and Cultural Organization (UNESCO) believes that between 10% and 20% of people worldwide suffer with dyslexia, indicating that the disorder is highly prevalent. It might potentially be as high as 30% in low-income areas. In traditional educational institutions, children with dyslexia frequently encounter obstacles that restrict their exposure to critical skills like programming, which are vital in the digital age. IT skills are a crucial tool to support dyslexic pupils because they can help them learn and reduce difficulties with identifying or confusing words and meaning. Several studies have demonstrated the importance of exposing students to IT skills, such as programming, especially those with special learning difficulties. Pupils with dyslexia typically have low expectations for their academic opportunities and low self-esteem. This talk examines how Python's ease of use and adaptability might be utilized to produce educational materials and programming tools that are accessible to dyslexic kids in underdeveloped countries. These resources are perfect for countries with limited resources because they require less infrastructure for technology. Join us as we explore python for inclusive education and creating a new generation of pupils who are masters of digital skills regardless of their situation and showing that coding is for everyone, breaking barriers and giving equal opportunities to dyslexic kids. pyladiescon-2025-78209-breaking-educational-barriers-for-pupils-with-dyslexia-using-python-in-africa Oluwakemi Omowunmi Jacobs en false https://pretalx.com/pyladiescon-2025/talk/A7MQRH/ https://pretalx.com/pyladiescon-2025/talk/A7MQRH/feedback/ Main Stream Panel 2025-12-06T10:55:00+00:00 10:55 00:30 Generative AI is everywhere, and it’s easy to get caught up in the excitement. In this session, our panelists will take a grounded look at how Python is being used to create real, practical GenAI projects. We’ll explore how the field started, what tools and experiments laid the groundwork, and where Python applications are making the biggest impact today. Panelists will share their experiences moving ideas from prototypes to meaningful projects, offer tips on essential skills and best practices, and discuss how to navigate ethical considerations and responsible AI use. Attendees will leave with practical insights, inspiration, and a clearer idea of how to turn curiosity about GenAI into real-world impact. pyladiescon-2025-84089-practical-genai-with-python-beyond-the-hype Jessica Greene (she/her)Jodie BurchellUbaida AbdulaiIulia FeroliUna Galyeva en false https://pretalx.com/pyladiescon-2025/talk/SPYPD3/ https://pretalx.com/pyladiescon-2025/talk/SPYPD3/feedback/ Main Stream Talk 2025-12-06T11:25:00+00:00 11:25 00:20 As a language teacher turned AI developer, I set out to create something that combines both of my passions: a self-hosted Swedish tutor powered by open-source models. This DIY project combines Ollama, LangGraph, and Gradio into a modular system that runs locally, delivers nuanced feedback, and costs nothing to run. No cloud inference, no API keys, just a playful, powerful way to decentralize AI and make language learning fun again. pyladiescon-2025-78580-building-a-self-hosted-ai-swedish-tutor-a-lovechild-of-teaching-and-tech Milena Miernik en At B2/C1 level, language learning resources tend to dry up, and Swedish is no exception. So in a true DIY spirit, I decided to build my own AI-powered tutor. In this talk, I’ll show how I combined my past life as an English teacher with my current role in AI software development to create a self-hosted, modular Swedish learning tool. You'll get a behind-the-scenes look at why I love Ollama and how I used it to run local models, LangGraph to build agentic AI workflows, and Gradio for a lightweight, interactive frontend. This talk isn’t just about language learning, it’s a case study in creative constraint, smart model selection, prompt engineering, and building usable AI systems with zero commercial cloud dependencies. If you're curious about multilingual LLMs, agentic reasoning, or sustainable AI development, this is for you. false https://pretalx.com/pyladiescon-2025/talk/JFDDUV/ https://pretalx.com/pyladiescon-2025/talk/JFDDUV/feedback/ Main Stream Talk 2025-12-06T11:45:00+00:00 11:45 00:20 Ever spent 40 minutes choosing a movie, only to rewatch something you didn’t even love? In this talk, I’ll share how I built Movie Groovy, a Python-based app that recommends movies based on how you feel. I’ll walk through how I used tools like FastAPI, Qdrant, Gemini (and some more) to create emotionally-aware suggestions, combining filtering logic, a lightweight Retrieval-Augmented Generation (RAG) approach, and an agentic flow that guides users step-by-step through their choices, for a more personalized experience. More than the stack, we’ll explore why empathy matters in AI, how user feedback shaped the design, and how you can build meaningful tools using open-source solutions, even as a solo dev. If you’re into Python, curious about AI beyond the hype, or just want to build things that feel more human, this talk is for you. pyladiescon-2025-78197-movie-groovy-building-a-mood-based-recommender-with-python-and-empathy Vassiliki Dalakiari en This talk shares the story behind Movie Groovy, a mood-based movie recommendation app I built using Python and lightweight GenAI techniques. Inspired by a common real-life frustration (trying to choose something to watch after a long day that actually matches how you feel), the project blends logic, emotional UX, and modern GenAI tools into a thoughtful user experience. I’ll walk through the high-level design choices, how I combined retrieval and generation in a simple but effective way, and the role of agentic design in shaping dynamic, emotionally-aligned results. I’ll share the key ideas, lessons learned, and the thinking process that made it work. You’ll leave with: - An idea of how GenAI can be used creatively in everyday tools - Ideas for integrating emotional context into user-facing apps - Inspiration to start or refine your own solo project - A look at how open-source tech can be used to build something meaningful: fast and with empathy This talk is suitable for anyone interested in Python, GenAI, or building applications that feel more human. false https://pretalx.com/pyladiescon-2025/talk/QQDM7H/ https://pretalx.com/pyladiescon-2025/talk/QQDM7H/feedback/ Main Stream Talk 2025-12-06T12:30:00+00:00 12:30 00:20 Sabemos que uma comunidade de código aberto não se resume apenas ao código, mas também às interações e conexões interpessoais. No entanto, como podemos medir e compreender o sentimento de pertença que nos torna mais confiantes para contribuir para o código aberto, após experienciarmos um ambiente verdadeiramente inclusivo? As emoções são tão importantes para a participação quanto as competências técnicas que os membro adquirem ao longo do tempo. Nesta palestra, apresentarei uma análise baseada em dados concretos sobre como a dinâmica emocional afeta o crescimento, a inclusão e a sustentabilidade dos capítulos das PyLadies. A partir de dados de inquéritos, experiências vividas e das melhores práticas da psicologia comunitária, abordarei: 1. O papel das emoções na construção da confiança, segurança e colaboração; 2. Como reconhecer os primeiros sinais de desmotivação e esgotamento emocional em voluntários (Baseado em simulações usando Python + LLMs - experiente em andamento); No final, espero trazer pontos de reflexão sobre as dinâmicas emocionais e seus influentes. pyladiescon-2025-79592-beyond-code-dinamicas-emocionais-das-comunidades-pyladies Cecilia Tivir pt false https://pretalx.com/pyladiescon-2025/talk/JSXHTQ/ https://pretalx.com/pyladiescon-2025/talk/JSXHTQ/feedback/ Main Stream Talk 2025-12-06T12:50:00+00:00 12:50 00:20 My personal perspectives from looking for a job as a FLINTA Python Engineer in 2025. pyladiescon-2025-79490-flinta-job-hunting-in-2025 Erica Wolf en I've been working in Python professionally for over 20 years. I have been building software for almost 40 years. I live in Berlin. I don't have a job. Why? Is it because of my gender (FLINTA)? Is it because of my hair color (purple FTW)? Is it because I am neuro-diverse (AudHD)? Is it because of my tattoos? While I can't get provide a better answer than what I get from recruiter, hiring managers, engineers reviewing my code assessments. I can share what I have done: (got the heck out of the USA). And I can make recommendations about what works (gets me interviews) and what doesn't. false https://pretalx.com/pyladiescon-2025/talk/AWCC7R/ https://pretalx.com/pyladiescon-2025/talk/AWCC7R/feedback/ Main Stream Talk 2025-12-06T13:15:00+00:00 13:15 00:20 Have you heard the expression, Losing the Forest Through the Trees? In a nutshell it means being so focused on details that we miss the bigger picture. Often in geospatial we are guilty as charged. It is about this tool or that platform vs the important work at hand. Building efficient workflows are an important part of building capacity to hold bigger stories. A powerful workflow I share when building data stories includes multi-faceted python libraries integrated with a wide variety of visual storytelling tools. Specifically QGIS, blender, Unreal engine and open source integration bringing a series of interconnected issues—including deforestation, pollution, habitat destruction, and biodiversity loss—representing a detachment from ecological systems, exacerbated by not only the energy requirements of exponential AI implementation but global industrial practices that allow environmental harm to be displaced geographically and socially. Quantitative Storytelling portrays an intricate dance between objective reality—Using the city as an operational unit we will build schema from multiple data sources and learn to tell stories about urban form and morphometrics learning from social demographics and the vulnerabilities encountered with climate risk and change. pyladiescon-2025-79894-decoding-climate-science-with-python-and-open-data-resources-what-ai-s-energy-needs-mean-for-our-planet bonny p mcclain en In an era dominated by pressing environmental challenges, the field of geospatial analysis holds the potential to illuminate critical narratives often obscured by technical details. This project explores the use of Python-based workflows and visual storytelling tools—specifically QGIS, Blender, and Unreal Engine—to create cohesive narratives that highlight interconnected ecological issues such as deforestation, pollution, habitat destruction, and biodiversity loss. These interconnected concerns illustrate the societal detachment from ecological systems, which is further complicated by the energy demands of AI technologies and prevailing industrial practices. Through a quantitative storytelling framework that leverages urban morphometrics and social demographics, we will reveal the vulnerabilities faced by communities in the context of climate risk. This effort seeks to transition the discourse from mere technicality to impactful narratives that drive awareness and inspire meaningful action towards environmental sustainability. false https://pretalx.com/pyladiescon-2025/talk/GMGLFN/ https://pretalx.com/pyladiescon-2025/talk/GMGLFN/feedback/ Main Stream Talk 2025-12-06T13:35:00+00:00 13:35 00:20 Stratified sampling helps ensure representative data subsets, especially when working with imbalanced groups. In this talk, we’ll explore its use in survey analysis, demonstrate Python-based implementation, and share best practices for improving data reliability and reducing bias. Attendees will gain a clear understanding of when and how to use stratified sampling effectively in real-world scenarios. pyladiescon-2025-79740-stratified-sampling-in-survey-research Cary Jim en Whether you're conducting surveys, building predictive models, or working with imbalanced datasets, stratified sampling is a powerful technique to ensure your data is representative and your insights are reliable. In survey research, especially within diverse populations like college students, simple random sampling can lead to underrepresentation of key subgroups. Stratified sampling addresses this by dividing the population into distinct strata (e.g., age groups, majors, demographics) and sampling proportionally from each. This approach reduces sampling bias and improves the reliability of statistical estimates. In this presentation, I will demonstrate how to use Python to implement stratified sampling in a student survey at our college. By maintaining proportional representation across age groups, we ensured that our sample reflected the diversity of the student body. I’ll walk through the conceptual foundations of stratified sampling, compare it with simple random sampling, and highlight its advantages in survey research and data analysis. We’ll also explore how to calculate weighted means when sample proportions differ from population proportions, which is a crucial step in producing unbiased estimates. Code examples and visualizations will be shared via GitHub to support hands-on learning and reproducibility. Learning Outcomes: 1. Understand when to use stratified sampling over random sampling 2. Learn how to define subgroups (strata) and implement sampling in Python 3. Evaluate sample representativeness by checking distribution patterns 4. Apply weights to adjust for unequal proportions in score estimation false https://pretalx.com/pyladiescon-2025/talk/GJNWXT/ https://pretalx.com/pyladiescon-2025/talk/GJNWXT/feedback/ Main Stream Talk 2025-12-06T13:55:00+00:00 13:55 00:20 MkDocs is a really simple tool to quickly create a static web site. I use it to quickly create web pages from basic markdown files. Recently I have also used Material for MkDocs to create a new version of my personal web page and it has been a really easy. If you do not have much time, MkDocs is a nice solution to quickly publish some web content. pyladiescon-2025-80303-creer-facile-un-site-web-avec-mkdocs-easily-create-a-website-with-mkdocs Françoise CONIL fr I like static site generators. I use Sphinx for a long time as I like reStructuredText. My technical notes (240 pages) are written in reStructuredText. I like to use MkDocs to quickly gather and publish some Markdown content that I may need to share as most people use Markdown. I am not a Markdown fan but MkDocs is simple to understand, to start with and to use. For the technical notes, I use Read the Docs theme. I recently dive into Material for MkDocs to recreate my personal web pages. I use Quarto to create slides. Our community has the chance to have many nice tools as we also have Pelican, Nicola, … false https://pretalx.com/pyladiescon-2025/talk/XGNKTB/ https://pretalx.com/pyladiescon-2025/talk/XGNKTB/feedback/ Main Stream Talk 2025-12-06T14:20:00+00:00 14:20 00:20 This novice-friendly project uses Python (plus Jekyll, a tried-and-true static site generator) to create an online cookbook that you and your friends can contribute to via email. I'll walk viewers through how I built this and why, and share source code for anyone interested in making their own cookbook. Attendees will get an overview of FastAPI, a handful of recipe parsing tools, and (if time allows) Jekyll. No knowledge of Jekyll is required, but you'll get the most out of this "how I built this" session if you have a basic understanding of Python. pyladiescon-2025-79231-let-s-build-a-cookbook-in-python Rachel Kaufman en false https://pretalx.com/pyladiescon-2025/talk/KMCPKM/ https://pretalx.com/pyladiescon-2025/talk/KMCPKM/feedback/ Main Stream Talk 2025-12-06T14:40:00+00:00 14:40 00:20 Can data engineering be feminist? Absolutely. As data engineers and Python developers, the decisions we make about what data to collect, how we transform it, and who gets access carry ethical weight. This talk explores how we can apply feminist principles like transparency, consent, and equity to build more ethical data pipelines. Through real-world examples of harm caused by opaque and biased data systems, we’ll examine how even routine ETL work can reinforce injustice. We’ll then outline practical approaches for designing pipelines that center care and accountability covering techniques like schema minimization, anonymization in Python, and documenting data provenance. Attendees will leave with a deeper understanding of how to challenge extractive data practices and use their technical skills to push back against surveillance capitalism and systemic bias. pyladiescon-2025-79553-feminist-pipelines-engineering-accountability-into-your-data-stack Yashasvi Misra en This talk begins by grounding the audience in the concept of feminist data engineering, drawing from data feminism, intersectional theory, and real-world cases where unethical data practices have caused harm such as biased recommendation systems or invasive tracking tools. We’ll examine how traditional data engineering workflows often prioritize efficiency, scale, and surveillance over consent, context, and care. From there, the talk transitions into a practical framework for ethical data design, introducing tools and techniques that Python engineers can use today. We’ll explore how to implement schema minimization, anonymization, and consent-aware collection in Python using libraries like Pandas and PyArrow. The talk also covers how to build transparency into ETL pipelines through audit logs and documentation practices. Throughout, I’ll include examples from my own experience working with legacy systems and modern pipelines to illustrate how ethical decisions show up in technical details. We’ll close by reflecting on how to advocate for ethical practices within teams and organizations, especially when trade-offs or pushback arise. This session is both a call to awareness and an invitation to action—for engineers who want to build systems that not only work, but do good. false https://pretalx.com/pyladiescon-2025/talk/HSUHCH/ https://pretalx.com/pyladiescon-2025/talk/HSUHCH/feedback/ Main Stream Keynote 2025-12-06T15:00:00+00:00 15:00 00:30 "We came for the code but stayed for the community." Tech is more than raising and merging PRs or clearing tasks on kanban boards—it is also about people, finding purpose, and belonging. In this keynote, we'll go through a journey, somewhat similar to yours—from gaining technical skills/mentorship to building a strong sense of community. We will talk about stories of growth, giving back to the community that waters, and creating meaningful connections that uplift others. pyladiescon-2025-82832-beyond-code-building-careers-growing-inclusive-communities Abigail Afi Gbadago en "We came for the code but stayed for the community." Tech is more than raising and merging PRs or clearing tasks on kanban boards—it is also about people, finding purpose, and belonging. In this keynote, we'll go through a journey, somewhat similar to yours—from gaining technical skills/mentorship to building a strong sense of community. We will talk about stories of growth, giving back to the community that waters, and creating meaningful connections that uplift others. false https://pretalx.com/pyladiescon-2025/talk/BUP8JK/ https://pretalx.com/pyladiescon-2025/talk/BUP8JK/feedback/ Main Stream Talk 2025-12-06T15:35:00+00:00 15:35 00:20 Many Python developers write code that works beautifully until it lands in production and something breaks. Suddenly, the bug isn’t in the logic, it’s in what you didn’t see coming. Without proper logging, metrics, or monitoring, it’s hard to answer even basic questions like: “Is this function working as expected?”, “Why did this job fail?”, or “How many users are affected?” This talk is for anyone who’s deployed (or wants to deploy) Python apps and has ever struggled to debug or monitor them after launch. I’ll introduce the concept of observability but without the buzzwords or complexity. Through real-life scenarios, I’ll explain how even a few lines of thoughtful logging and some basic metrics can transform one’s code from “it works on my machine” to “I can debug this at 2 AM.” We’ll walk through examples of poor vs. effective logging, how metrics reveal patterns logs can’t, and how monitoring ties it all together. Whether one is building web apps, CLIs, or automation scripts, this talk will show how one can start writing more observable Python. pyladiescon-2025-79484-the-python-you-didn-t-write-understanding-logs-metrics-and-monitoring-for-better-production-code Lois Bassey en ⸻ Why This Talk? Most developers focus on writing code that works. But production environments are full of unknowns: bad data, service outages, weird edge cases. What helps a developer survive these moments isn’t just better code, It’s better visibility. My talk addresses a knowledge gap many Python developers face: how to make their code observable, so they can understand, troubleshoot, and improve it, especially once it’s deployed. ⸻ Key Takeaways: By the end of this talk, attendees will: • Understand what observability means and why it matters. • Know the difference between logs, metrics, and monitoring, and when to use each. • See what good logging looks like (and what to avoid). • Walk away with actionable ideas for improving visibility into their code (whether it’s a web app, script, or background worker). false https://pretalx.com/pyladiescon-2025/talk/R9NZ7G/ https://pretalx.com/pyladiescon-2025/talk/R9NZ7G/feedback/ Main Stream Talk 2025-12-06T15:55:00+00:00 15:55 00:20 I’m an experienced data analyst consultant working with data in international development, but I only started learning Python less than a year ago. As a beginner programmer with a passion for data visualization design, I began exploring how to build interactive dashboards with Python and Dash — not just to keep coding, but to share stories that matter deeply to me. In this talk, I share key lessons learned so far — through a beginner’s lens — the shift from a rough idea to a working concept, choosing data, structuring the dashboard, and finding the balance between clarity and aesthetics in my design decisions. I’ll focus on one dashboard project as a case study to show what worked, what didn’t, and how I adapted along the way. Specifically, I will focus on correlations between malaria risk in children under 5 in Nigeria and other key indicators such as climate change and access to water and sanitation. I’ll also share my personal experience of troubleshooting when nothing seems to work, and how I kept motivation up when hope was dim. My goal is to encourage others to start their own “imperfect” projects, embrace iteration, and see learning as an ongoing process with both highs and lows. If you’re curious about Python’s capabilities for data storytelling, want to explore interactive dashboards, or simply need that pep talk to begin your first Python project, this talk will offer insights, encouragement, and — most likely — a glimpse of your own journey reflected back at you. pyladiescon-2025-79690-from-fog-to-lighthouse-a-beginner-s-journey-from-raw-data-to-a-functional-dashboard Loredana Horezeanu en My talk revolves around building interactive dashboards with Python and Dash — from the perspective of someone who is still a beginner in Python programming, but experienced in working with complex data, other programming languages like R, and creating impactful visualizations in the field of international development. I’ve spent over a decade as a data analyst consultant, translating global health, education, and development data into stories for decision-makers. Less than a year ago, I began learning Python, motivated by two goals: to expand my technical skills and to explore how interactive dashboards can make complex global issues both accessible and compelling. From the start, I knew I wanted to tell stories that matter deeply to me. For this talk, I will focus on one dashboard concept — for example, exploring the intersection of WASH, climate and malaria risk in children under 5 in Nigeria — as a case study. This narrower scope will allow me to dig deeper into the process, decisions, and lessons learned. I’ll share: How I went from a raw idea to a working concept Choosing datasets that are both technically workable and narratively rich Structuring dashboards so users can follow the story without getting lost Striking the balance between clarity and aesthetics in design Identifying what felt surprisingly straightforward and what turned into roadblocks I’ll also speak openly about the challenges — app architecture, stubborn bugs, complex callbacks, messy data — and the moments where I had to rethink my approach entirely when nothing else worked. Alongside the technical side, I’ll cover the emotional reality of learning something new: how to keep going when progress is too slow, how to find motivation in small wins, and how to stay connected to the bigger “why” behind the project. This talk is not about providing the audience with a flawless final product — it’s about sharing an honest, evolving journey. Attendees will leave with realistic expectations of learning Dash as a Python beginner, strategies for turning ideas into prototypes, and encouragement to start their own “imperfect” projects. false https://pretalx.com/pyladiescon-2025/talk/JWQWJ9/ https://pretalx.com/pyladiescon-2025/talk/JWQWJ9/feedback/ Main Stream Talk 2025-12-06T16:15:00+00:00 16:15 00:20 Esta charla aborda de manera introductoria la aplicación de la ética en la IA, específicamente en términos de interpretabilidad y cómo la opacidad de estos modelos genera riesgos tanto técnicos como sociales en un contexto de adopción acelerada. Analizaremos casos reales donde la falta de transparencia en los procesos de toma de decisiones ha llevado a consecuencias inesperadas. Además, veremos cómo los sistemas que han traído el auge comercial a la inteligencia artificial (los modelos de lenguaje natural) cuentan con mecanismos ocultos que, afortunadamente, se han comenzado a analizar. Con esto, la audiencia podrá conformar una opinión sobre qué tanto hay detrás de cada una de las respuestas que parecen coherentes y neutrales. Finalmente, dentro de la charla exploraremos cómo los sistemas de IA se han convertido en infraestructuras sociotécnicas que requieren perspectivas multidisciplinarias, y cómo nosotros, como usuarios o desarrolladores, tenemos cabida desde una perspectiva más analítica y participativa en este ecosistema. pyladiescon-2025-80275-etica-en-la-ia-un-espacio-comun-para-construir-en-conjunto Karla Angelica Doctor Mauricio es Los sistemas de inteligencia artificial trascienden el ámbito técnico para convertirse en sistemas sociotécnicos complejos que requieren perspectivas multidisciplinarias para identificar sesgos en datos y correlaciones espurias que pueden generar resultados poco interpretables. El auge exponencial de la IA ha creado una percepción de necesidad competitiva que lleva a empresas, instituciones e individuos a integrarla aceleradamente en sus procesos. Esta adopción masiva genera riesgos significativos, especialmente cuando la falta de interpretabilidad y transparencia en los modelos genera una confianza ciega que puede superar al conocimiento empírico y la evaluación crítica. En esta charla abordaremos: Interpretabilidad como desafío central: Análisis de cómo la opacidad en los procesos de toma de decisiones de los modelos genera riesgos éticos y operacionales Casos reales de falta de transparencia: Ejemplos concretos donde la imposibilidad de explicar las decisiones de la IA ha llevado a consecuencias inesperadas y problemáticas Mecanismos ocultos de razonamiento: Vislumbrar cómo funcionan los modelos detrás de sus respuestas aparentemente coherentes y neutrales Ciclos de construcción problemáticos: Cómo la falta de explicabilidad en los modelos amplifica sesgos técnicos y éticos Recomendaciones prácticas: Consideraciones y acciones concretas para un uso más consciente y responsable de estas tecnologías false https://pretalx.com/pyladiescon-2025/talk/ZQ98FN/ https://pretalx.com/pyladiescon-2025/talk/ZQ98FN/feedback/ Main Stream Keynote 2025-12-06T17:00:00+00:00 17:00 00:30 ## Abstract **Don't Be Everything to Everyone: The Case for Career and Model Specialization (Small Language Models)** We’ve all heard the advice: “Be versatile, keep your options open.” But here’s the twist — whether you’re building a career or building a language model, trying to be everything to everyone is a recipe for burnout and bloat. The real magic happens when you specialize. This keynote makes parallels between professional growth and AI design: - 🎯 **Sovereignty of your stack**: Just like owning your career path, owning your open-source tools gives you freedom, resilience, and a little swagger. - 🔄 **Small iterations, big wins**: Careers aren’t built in one giant leap, and neither are models. Iteration is the secret sauce that compounds into mastery. - ⚡ **Efficiency through expertise**: Specialists — human or machine — don’t waste energy chasing every problem. They focus, optimize, and deliver outsized impact. Think of it as a pep talk for both your résumé and your repo. By the end, you’ll see why the future belongs to those who go deep, not wide — and why small, sharp models (and careers) punch way above their weight. pyladiescon-2025-84760-don-t-be-everything-to-everyone-the-case-for-career-and-model-specialization-small-language-models Dawn Wages en Why specialize? Because neither your résumé nor your language model should try to do everything. This keynote makes the case for going small and going deep — showing how sovereignty, iteration, and expertise create careers (and models) that not only meets the brief, but helps uncover innovations in your career and your domain analysis. Think of it as a pep talk for both your résumé and your repo. It draws on my personal experience as a consultant, software engineer, analyst, developer relations engineer, manager, self-taught programmer and self-proclaimed "jack of all trades" -- now deeping expertise in data science and machine learning, where I can share what I've learned in the value of specialization. false https://pretalx.com/pyladiescon-2025/talk/DUU7CV/ https://pretalx.com/pyladiescon-2025/talk/DUU7CV/feedback/ Main Stream Panel 2025-12-06T17:30:00+00:00 17:30 00:30 Este panel reúne a cuatro mujeres del mundo tech para compartir, en una conversación honesta y práctica, los principales retos y oportunidades que han enfrentado en sus trayectorias profesionales. A lo largo del diálogo explorarán decisiones técnicas complejas, momentos clave que marcaron sus carreras, tensiones entre ética y plazos del negocio, y los desafíos de crecer en una industria en constante cambio. También abordarán las oportunidades actuales, el impacto de la comunidad en la construcción de pertenencia y el contexto real de inclusión en 2025. El panel ofrece perspectivas reales, acciones concretas y consejos prácticos para quienes buscan avanzar, reinventarse o encontrar su lugar en tecnología. pyladiescon-2025-84090-retos-y-oportunidades-en-el-mundo-tech Nicole Franco LeonKarina Fernanda PérezCarla Marcela Florida RománMaría-Fernanda Villalobos - ModeratorLiliana Garmendia Campos es Este panel reúne a cuatro mujeres del mundo tech para compartir, en una conversación honesta y práctica, los principales retos y oportunidades que han enfrentado en sus trayectorias profesionales. A lo largo del diálogo explorarán decisiones técnicas complejas, momentos clave que marcaron sus carreras, tensiones entre ética y plazos del negocio, y los desafíos de crecer en una industria en constante cambio. También abordarán las oportunidades actuales, el impacto de la comunidad en la construcción de pertenencia y el contexto real de inclusión en 2025. El panel ofrece perspectivas reales, acciones concretas y consejos prácticos para quienes buscan avanzar, reinventarse o encontrar su lugar en tecnología. false Retos y oportunidades en el mundo tech https://pretalx.com/pyladiescon-2025/talk/CLYKNU/ https://pretalx.com/pyladiescon-2025/talk/CLYKNU/feedback/ Main Stream Talk 2025-12-06T18:05:00+00:00 18:05 00:20 En esta charla mostraré cómo Python, gracias a sus librerías de código abierto, permite pasar rápidamente de un conjunto de datos con coordenadas a un mapa visual y atractivo. Usaremos ejemplos prácticos con herramientas como folium, geopandas y plotly.express para crear desde mapas simples hasta visualizaciones interactivas. La idea es que cualquier persona, incluso sin experiencia previa en GIS, pueda entender y aplicar estos conceptos para contar historias con datos geográficos. pyladiescon-2025-79696-de-coordenadas-a-colores-visualizando-el-mundo-con-python Ingrid Solis es Python no solo es un lenguaje versátil para el análisis de datos, sino también una poderosa puerta de entrada al mundo de la visualización geoespacial. En esta charla exploraremos cómo, gracias a su ecosistema de librerías de código abierto, es posible crear desde mapas estáticos simples hasta visualizaciones interactivas y dinámicas que se pueden compartir fácilmente. Usaremos herramientas como geopandas, folium y plotly.express para trabajar con distintos formatos de datos geográficos, desde shapefiles y GeoJSON hasta archivos CSV con coordenadas. Partiremos desde cero: cargando datos, procesándolos y llevándolos a un mapa, para luego incorporar elementos visuales como colores, capas, pop-ups y filtros interactivos. También revisaremos fuentes abiertas de datos geoespaciales que permiten enriquecer nuestras visualizaciones, como OpenStreetMap o portales de datos públicos. El objetivo es que cualquier persona, sin importar su experiencia previa en GIS, pueda entender el flujo completo: desde recibir un dataset en bruto hasta entregar un mapa que cuente una historia y facilite la toma de decisiones en áreas tan diversas como análisis urbano, logística, medio ambiente o investigación social. Al final, verás que pasar del dato al insight geográfico con Python puede ser tan simple como escribir unas pocas líneas de código. false https://pretalx.com/pyladiescon-2025/talk/WZTMVK/ https://pretalx.com/pyladiescon-2025/talk/WZTMVK/feedback/ Main Stream Talk 2025-12-06T18:45:00+00:00 18:45 00:20 Too often, we run Python workshops, teach people the basics, and then… leave. And while those workshops are exciting and inspiring, they’re not enough. In this talk, I’ll share lessons from years of teaching and what I learned organizing Django Girls Koforidua. I’ll walk through practical ways to move beyond one-off workshops to building real learning experiences: supportive environments, hands-on projects, and community-based follow-ups that help learners grow and stay engaged. Whether you’re an educator, developer, or community organizer, you’ll leave with ideas for how to make your next workshop the beginning of a learning journey—not the end of one. pyladiescon-2025-79525-workshops-are-a-start-not-a-path Barbara Asiamah en Workshops are a good start, but they’re not enough. If we want people to really learn Python, we have to build experiences that support them after the workshop—resources, projects, and community that help them keep going. Let’s stop teaching and disappearing false https://pretalx.com/pyladiescon-2025/talk/DL9FEZ/ https://pretalx.com/pyladiescon-2025/talk/DL9FEZ/feedback/ Main Stream Talk 2025-12-06T19:10:00+00:00 19:10 00:20 Acquisition of satellite imagery, frequently becomes corrupted by various forms of noise, which degrades image quality and impedes accurate interpretation and subsequent analysis. This paper presents a novel, integrated approach for effective noise removal from such imagery, utilizing the capabilities of the Wavelet transform based thresholding, Singular Value Decomposition (SVD), and also autoencoders. The Wiener filter, proves adept at addressing stationary noise, particularly effective in suppressing additive Gaussian noise whilst preserving important details based on local statistics. Following this, SVDs are employed to exploit the inherent low-rank structure characteristic of natural images, enabling further noise reduction by reconstructing the image from its more significant singular values, thereby separating signal from noise components. Finally, an autoencoder, which is are powerful deep learning architecture, is then utilized for its capacity to learn complex, non-linear representations from noisy inputs. This data-driven strategy allows the autoencoder to effectively apprehend intricate noise patterns and reconstruct high-fidelity images, offering particular benefit for complex or non-stationary noise types where traditional filters may struggle. By combining these complementary techniques, the proposed methodology aims for superior noise suppression, the preservation of crucial image features, and the enhancement of overall visual quality and analytical utility of satellite images. Expected results demonstrate significant improvements in signal-to-noise ratio and perceptual quality compared to individual methods, highlighting the synergistic benefits of this combined framework. pyladiescon-2025-79236-a-low-cost-automatic-compressing-and-noise-removing-algorithm-for-satellite-images SREYA GHOSH en We have demonstrated a preliminary approach to solving the satellite denoising problem. Wavelet, as we know, is an excellent tool in studying signals, especially time-dependent data like that of satellite imagery. We have used a Weiner filter and applied wavelet and SVD in a hybrid algorithm and studied it its development so far, and how good it is serving to denoise the data. In parallel, we are training our autoencoder with the data to produce an algorithm efficient enough to denoise satellite data and computationally efficient! false https://pretalx.com/pyladiescon-2025/talk/UAJAKB/ https://pretalx.com/pyladiescon-2025/talk/UAJAKB/feedback/ Main Stream Talk 2025-12-06T19:30:00+00:00 19:30 00:20 Behind every great data product is a perfectly balanced recipe of tools and methodologies, and Python is the master chef that can bring them all together. Whether you're a data engineer, analyst, or developer, you’ll walk away with a practical blueprint for building scalable, modern data workflows with just the right ingredients, and no risk of burning the cake. pyladiescon-2025-79750-python-powered-data-engineering-how-to-build-a-modern-end-to-end-stack Giulia Silvestro en Today’s data teams are under pressure to deliver insights quickly, accurately, and at scale, without getting lost in complexity. Python makes this possible bringing ingestion, transformation, orchestration, and visualization together in a single, streamlined workflow. In this talk, we’ll walk through building a complete sales analytics pipeline for an e-commerce business, showing how Python’s rich ecosystem enables rapid, reliable, and scalable data delivery. The Pipeline Journey: 1. Ingestion Extract raw sales data from multiple source, using Python’s powerful data-access libraries. Validate and clean it for smooth downstream processing. 2. Storage Store your data in File Storage + DuckDB, a high-performance OLAP database that integrates seamlessly with Python, offering enterprise-grade analytics. 3. Transformation Combine dbt for modular SQL transformations with Polars for lightning-fast DataFrame operations. Model essential KPIs with accuracy and speed. 4. Orchestration Automate with Dagster or Airflow. We’ll explore DAG structure, dependency handling, and robust error recovery. 5. Visualization Present insights in an interactive Streamlit dashboard, where users can filter by product, channel, or date and see results instantly. Key Takeaways: - A proven blueprint for modern Python-powered data pipelines. - A production-ready reference architecture you can adapt immediately. - Practical guidance on choosing the right tools for your team and workload. false https://pretalx.com/pyladiescon-2025/talk/3RFQUE/ https://pretalx.com/pyladiescon-2025/talk/3RFQUE/feedback/ Main Stream Keynote 2025-12-06T20:20:00+00:00 20:20 00:30 We live in a golden age of programming paradigms and frameworks, all promising correctness, reproducibility, and even provability. And yet anyone who ships production code will, on a long enough timeline, become a connoisseur of failure. This is doubly true in tech-for-good initiatives, where the true complex system is a constrained society that doesn’t care how polished your code is. In this talk, I remix Richard Cook’s How Complex Systems Fail into a tour of unintended consequences, questionable decisions (mostly mine), and lessons learned, forgotten, and learned again over 20 years of trying to make the world better with computers. I hope you’ll leave with fewer “root cause” fairy tales, more humane failure modes in life and work, and maybe a little less fear of pressing deploy on your dreams. pyladiescon-2025-85073-how-complex-systems-taught-me-to-fail Imogen Wright en false https://pretalx.com/pyladiescon-2025/talk/QMHVMV/ https://pretalx.com/pyladiescon-2025/talk/QMHVMV/feedback/ Main Stream Talk 2025-12-06T20:50:00+00:00 20:50 00:20 Python is one of the most widely used programming languages in the world, but understanding how it works under the hood can be a challenge. Even though it is open source, there are very few accessible resources to explore CPython—its reference implementation in C—and even fewer if English is not your first language. But what if, instead of writing `if`, we could write `si`? What if `print()` became `imprimir()`? And what if, while transforming Python into Spanish, we learned how Python itself is built? In this talk, we’ll explore what CPython is and how to compile it, why Python is written in C instead of Python, and how its grammar and parser work. Finally, we’ll put this knowledge into action by changing Python’s keywords into Spanish—`if` becomes `si`, `for` becomes `para`, and `while becomes `mientras`. By learning how CPython works internally, you’ll gain a deeper understanding of why certain things work (and why some errors happen), experiment with new language features, create your own extensions, and maybe even contribute back to CPython itself. And although it may sound intimidating, CPython is not an untouchable project—you’ll see that it’s more approachable than you think. While the examples in this talk are in Spanish, the approach can be adapted to *any* natural language, including **yours**! This talk is aimed at Python users with intermediate knowledge who are curious about how the language is implemented. No prior experience with C, compilers, or language theory is required—just a willingness to learn something new. pyladiescon-2025-80604-how-to-teach-your-language-to-python-with-cpython Sofia Toro en false https://pretalx.com/pyladiescon-2025/talk/SNLZFB/ https://pretalx.com/pyladiescon-2025/talk/SNLZFB/feedback/ Main Stream Talk 2025-12-06T21:10:00+00:00 21:10 00:20 Ever wondered what happens to your data after you delete your account? While the database might forget you, the AI models trained on your data still "remember" your patterns. This talk introduces machine unlearning - the fascinating field of teaching AI to truly forget. In this beginner-friendly session, you'll discover why machine unlearning matters more than ever. From GDPR's "right to be forgotten" to removing bias from hiring algorithms, the ability to make AI forget specific information is crucial for privacy, fairness, and safety. I'll demonstrate some practical and theoretical Python approaches you can implement today: 1. SISA Framework - Split training data into shards, making forgetting 5x faster than retraining fromscratch 2. Gradient Ascent - Teach models to be confidently wrong about specific data 3. Strategic Noise Injection - Add controlled noise to confuse AI about unwanted memories I'll explain the key concepts and walk through Python examples, showing how different unlearning approaches work in practice. While some advanced techniques exist, I'll focus on methods that are approachable for those getting started with machine learning. Whether you're building your first ML model or working on production systems, understanding machine unlearning helps you create more ethical and privacy-conscious AI. You'll leave with practical code examples, resources for further learning, and a new perspective on responsible AI development. You just need curiosity about making AI more trustworthy and accountable. pyladiescon-2025-79452-main-character-energy-taking-back-control-of-your-data-from-ai Samiksha Khadka en What You'll Learn: - Why deleting data from databases doesn't remove it from AI models - Real-world scenarios where machine unlearning is essential (GDPR compliance, bias removal, security) - Hands-on Python techniques for implementing unlearning - How to build privacy-conscious ML systems from the start Technical Content: - This talk bridges theory and practice, making machine unlearning concepts accessible to Python developers at various skill levels. Attendees will gain understanding of when and how to apply these techniques. Who Should Attend: - Beginners curious about AI ethics and privacy - Python developers interested in machine learning - Anyone building ML systems who wants to handle data responsibly - Practitioners seeking GDPR-compliant AI solutions Key Takeaways: - Understand the difference between data deletion and model forgetting - Implement different unlearning approaches in Python - Recognize when and why to use machine unlearning in your projects - Access resources and code examples for continued learning false Ctrl + z for machine : The Art of Machine Unlearning https://pretalx.com/pyladiescon-2025/talk/VN37NU/ https://pretalx.com/pyladiescon-2025/talk/VN37NU/feedback/ Main Stream Talk 2025-12-06T21:35:00+00:00 21:35 00:20 Medical AI models often fail in African settings due to hidden dataset and model biases. This talk shows how to detect and measure bias in medical imaging using Python. We will explore real examples with MRI and X-ray data, learn subgroup performance analysis, apply fairness improvements, and discuss practical lessons for building more reliable and equitable healthcare AI. pyladiescon-2025-79571-why-your-medical-ai-model-might-not-work-in-africa-a-python-guide-to-measuring-bias Ilerioluwakiiye Abolade en Bias in AI is not just a fairness buzzword, it is a clinical safety issue. In this talk we will work through practical Python workflows for uncovering hidden biases in medical imaging models. You will learn: • How to explore dataset composition and label distribution with Pandas and Seaborn • How to compare subgroup performance using MONAI and TorchMetrics • Lightweight bias mitigation strategies, including reweighting and LoRA based fine tuning • What fairer performance actually looks like in practice using African brain MRI and chest X-ray examples I will share successes, failures, and unexpected findings from evaluating segmentation models across datasets. You do not need a medical background, only curiosity and some Python experience. You will leave with practical tools to ensure your models work more equitably for diverse patient populations, especially within the African context. false https://pretalx.com/pyladiescon-2025/talk/K9QRPW/ https://pretalx.com/pyladiescon-2025/talk/K9QRPW/feedback/ Main Stream Talk 2025-12-06T21:55:00+00:00 21:55 00:20 The classic open source dichotomy is well known: everyone uses it, but few give back for the convenience we've all grown so used to. What sustains a community? And how can we help ensure that the frameworks and libraries we rely on daily remain viable and healthy in the long run? In this talk, I intend to show how easy it is to get started with your open source contributions, regardless of your seniority. pyladiescon-2025-79689-what-if-you-became-an-open-source-software-contributor Amanda Savluchinske en The classic open source dichotomy is well known: everyone uses it, but few give back for the convenience we've all grown so used to. What sustains a community? And how can we help ensure that the frameworks and libraries we rely on daily remain viable and healthy in the long run? Contributing is easier than it seems. Maintainers are, more often than not, more welcoming and kind than you'd expect, and I want to show you how simple it can be to take your first steps in this world. After all, it wasn't that long ago that I was in that exact same place. In this talk, you can expect to: - Learn a bit more about the open source ecosystem; - Understand how these decentralized communities operate at a basic level; - Get practical tips on how to start contributing and explore possible paths you can take. Talk outline: - Personal introduction - Open source and its sustainability - Demystifying barriers to contributing - Maintainers — what do they want? - How senior do I need to be to contribute? - How to start contributing? - Mentorship programs - Djangonaut Space - Getting involved in communities and forums - Practical tips for contributing to OSS projects - Navigating repositories - Using AI to help understand codebases - Finding good issues to start contributing - Best practices for contributing - Conventional Commits - TDD - Documentation updates - Communicating with maintainers false https://pretalx.com/pyladiescon-2025/talk/DA7ZYQ/ https://pretalx.com/pyladiescon-2025/talk/DA7ZYQ/feedback/ Main Stream Talk 2025-12-06T22:15:00+00:00 22:15 00:20 本場分享將回顧講者自 2016 年起在 PyLadies Taiwan 的經驗，從擔任 Host 到退居幕後，逐步累積近十年的實踐與觀察，內容將聚焦在： 1. 如何在不同發展方向間取得平衡、決定社群成長的步調與樣貌，鼓勵更多志願者走上領導之路。 2. 如何在細節裡守護活動品質與參與者體驗。 希望透過這些實際案例，給予正在經營或參與社群的你一些靈感：如何長期維護一個技術社群，更能不斷培育新能量，在世代交替間持續茁壯。 pyladiescon-2025-80327-host--pyladies-taiwan Mars Weng zh-hant false https://pretalx.com/pyladiescon-2025/talk/QZFBCH/ https://pretalx.com/pyladiescon-2025/talk/QZFBCH/feedback/ Main Stream Panel 2025-12-06T22:40:00+00:00 22:40 00:30 Join our panelist to discuss an insightful and interesting topic! pyladiescon-2025-84091-de-10-a-100-em-lideranca-comunitaria Ana Cecília Vieira - ModeradoraJuliany RaiolJuliana Karoline de SousaFernanda Farias en false https://pretalx.com/pyladiescon-2025/talk/DU78CA/ https://pretalx.com/pyladiescon-2025/talk/DU78CA/feedback/ Main Stream Keynote 2025-12-06T23:10:00+00:00 23:10 00:30 Ser mujer en tecnología no es solo ocupar un espacio en un sector históricamente dominado por hombres: es transformar la manera en que lideramos, innovamos y construimos futuro. En esta charla exploraremos cómo el liderazgo femenino puede convertirse en una fuerza catalizadora dentro de la industria del software, impulsando culturas más humanas, colaborativas y de alto desempeño. Durante 30 minutos recorreremos los desafíos y oportunidades de liderar en el mundo tech desde la autenticidad, la visión estratégica y la empatía. Hablaremos de herramientas prácticas para desarrollar liderazgo técnico y organizacional, gestionar talento diverso, fomentar la innovación y construir equipos autónomos que escalan sin perder su esencia. También abordaremos las claves para crecer como líder: cómo navegar entornos de alta complejidad, ganar influencia sin perder la voz propia, y convertir la diversidad en una ventaja competitiva. A través de ejemplos reales, compartiré aprendizajes sobre cómo crear entornos donde las personas puedan brillar, las ideas fluyan y la excelencia se sostenga en el tiempo. Esta charla es una invitación a repensar el liderazgo desde el ADN femenino: más colaborativo que jerárquico, más orientado al propósito que al control, y profundamente consciente de que la tecnología no solo resuelve problemas, sino que moldea la forma en que habitamos el mundo. pyladiescon-2025-82374-liderazgo-femenino-en-el-mundo-tech Patty Trejo es false https://pretalx.com/pyladiescon-2025/talk/WRPAMN/ https://pretalx.com/pyladiescon-2025/talk/WRPAMN/feedback/ Main Stream Message from Organizers 2025-12-06T23:45:00+00:00 23:45 00:15 The conference is getting to an end, so join us for a retrospective discussion and a discussion of future versions! pyladiescon-2025-84092-conference-closing en false https://pretalx.com/pyladiescon-2025/talk/83GVBB/ https://pretalx.com/pyladiescon-2025/talk/83GVBB/feedback/ Activities & Open Spaces Workshop 2025-12-06T18:00:00+00:00 18:00 01:30 In this hands-on lab, you will build an agent-driven Retrieval-Augmented Generation (RAG) application using a U.S. Case Law dataset. You will work with Azure Database for PostgreSQL, Visual Studio Code, and the Semantic Kernel Agent Framework to explore AI-powered text and vector search. The lab also introduces GraphRAG with Apache AGE, enabling enriched querying across structured and graph-based data. By the end, you'll gain practical skills in semantic search, vector indexing, and intelligent data retrieval. pyladiescon-2025-85216-build-agentic-ai-with-semantic-kernel-and-graph-rag-on-postgresql Dingding LuGauri Kasar en false https://pretalx.com/pyladiescon-2025/talk/QSJHYR/ https://pretalx.com/pyladiescon-2025/talk/QSJHYR/feedback/ Activities & Open Spaces Workshop 2025-12-06T20:00:00+00:00 20:00 01:30 This hands-on lab guides you through the process of building a modern, full-stack web application using FastAPI for the backend and open source DocumentDB as the database, all orchestrated with Docker containers. Attendees will learn how to rapidly prototype and deploy scalable Python APIs, leverage containerization for consistent development and testing, and integrate with DocumentDB for flexible, cloud-ready data storage. In this session, you will learn: - How to set up a FastAPI backend and connecting it to DocumentDB using Docker Compose. - Best practices for local development, testing, and CI/CD integration with Docker. - How open source DocumentDB enables rapid iteration and seamless migration to cloud environments. - Real-world patterns for building, running, and debugging full-stack apps in isolated, reproducible environments. By the end of the lab, you’ll have a working application, a deeper understanding of containerized development workflows, and practical experience with the latest open source tools for Python and NoSQL development. pyladiescon-2025-85326-building-a-full-stack-fastapi-app-with-documentdb Patty Chow en false https://pretalx.com/pyladiescon-2025/talk/YJBN9H/ https://pretalx.com/pyladiescon-2025/talk/YJBN9H/feedback/