{"$schema": "https://c3voc.de/schedule/schema.json", "generator": {"name": "pretalx", "version": "2026.1.1"}, "schedule": {"url": "https://pretalx.com/open-source-ai-workshops-2026/schedule/", "version": "0.6", "base_url": "https://pretalx.com", "conference": {"acronym": "open-source-ai-workshops-2026", "title": "Open Source AI Workshops 2026", "start": "2026-05-18", "end": "2026-05-19", "daysCount": 2, "timeslot_duration": "00:05", "time_zone_name": "Europe/Berlin", "colors": {"primary": "#0e3a4c"}, "rooms": [{"name": "Raum A", "slug": "5271-raum-a", "guid": "40df791a-55b7-5cae-9a72-1a7c7269dd91", "description": null, "capacity": 30}, {"name": "Raum B", "slug": "5272-raum-b", "guid": "b4ccb075-a2c8-5ede-a2ae-05f758b0e9cb", "description": null, "capacity": null}, {"name": "Raum C", "slug": "5273-raum-c", "guid": "43b26fce-ba76-57c8-ad06-6124418ad758", "description": null, "capacity": null}, {"name": "Raum D", "slug": "5274-raum-d", "guid": "50e2a0ab-a107-5929-9198-08f6cb31f6be", "description": null, "capacity": null}, {"name": "Raum E", "slug": "5275-raum-e", "guid": "821251f2-d4cb-51f2-bced-bef91bb558f7", "description": null, "capacity": null}, {"name": "Raum F", "slug": "5276-raum-f", "guid": "cf47361f-a7f1-5e59-b930-015039183e26", "description": null, "capacity": 30}], "tracks": [], "days": [{"index": 1, "date": "2026-05-18", "day_start": "2026-05-18T04:00:00+02:00", "day_end": "2026-05-19T03:59:00+02:00", "rooms": {"Raum A": [{"guid": "7a4676a6-d655-549a-8aaa-df17d718d097", "code": "EBTWDL", "id": 88777, "logo": null, "date": "2026-05-18T09:00:00+02:00", "start": "09:00", "duration": "08:00", "room": "Raum A", "slug": "open-source-ai-workshops-2026-88777-einfuhrung-zu-open-source-ai", "url": "https://pretalx.com/open-source-ai-workshops-2026/talk/EBTWDL/", "title": "Einf\u00fchrung zu Open Source AI", "subtitle": "", "track": null, "type": "Workshop", "language": "de", "abstract": "In diesem Workshop werden die Grundlagen von AI vermittelt. Wie funktioniert (Generative) KI, was sind Transformer Modelle und wie werden basierend darauf Large Language Modelle (LLMs) entwickelt.\r\nDer Workshop versucht so gut wie m\u00f6glich ohne Mathematik auszukommen, sondern die Konzepte so weit zu vermitteln, dass ein Verst\u00e4ndnis daf\u00fcr entsteht.\r\n\r\nIn einem zweiten Teil wird vermittelt wie LLMs auf lokalen Ger\u00e4ten unter Einhaltung des Datenschutzes angewendet werden k\u00f6nnen. Dies machen wir anhand des Beispiels von LM-Studio, welches die Verwendung von LLMs auf lokalen Ger\u00e4ten erm\u00f6glicht.", "description": "Folgende Themen werden behandelt:\r\n\r\n- Was sind Transformer-Modelle und wie funktionieren sie?\r\n- Wie funktioniert Generative KI?\r\n- Lokale Nutzung von LLMs mittels ollama / LM-Studio\r\n- Wie funktioniert Retrieval Augmented Generation (RAG)?\r\n- Wie baue ich einen RAG-Chatbot auf meinen eigenen Dokumenten mit Open Source AI?\r\n\r\nZus\u00e4tzliche Diskussionsthemen, die im Rahmen des Workshops auf Wunsch angesprochen werden k\u00f6nnen:\r\n- Ethische Aspekte von KI (Bias, Fake News, Verantwortung)\r\n- Open-Source vs. Closed-Source KI-Modelle\r\n- Die Zukunft von LLMs und personalisierter KI", "recording_license": "", "do_not_record": false, "persons": [{"code": "B8HKNC", "name": "Gygli Marcel", "avatar": "https://pretalx.com/media/avatars/B8HKNC_4VmNYGp.webp", "biography": "Prof. Dr. Marcel Gygli is passionate about advancing Artificial Intelligence in the Public Sector. His focus is to drive transformative innovation, bringing cutting-edge AI research to practical applications in government.", "public_name": "Gygli Marcel", "guid": "24f8842e-62fa-5030-9f49-fe233e7bfe81", "url": "https://pretalx.com/open-source-ai-workshops-2026/speaker/B8HKNC/"}], "links": [], "feedback_url": "https://pretalx.com/open-source-ai-workshops-2026/talk/EBTWDL/feedback/", "origin_url": "https://pretalx.com/open-source-ai-workshops-2026/talk/EBTWDL/", "attachments": []}], "Raum B": [{"guid": "7f498bea-4fce-5cb6-9ee1-cfa9b1a789cc", "code": "HY8GHW", "id": 88790, "logo": null, "date": "2026-05-18T09:00:00+02:00", "start": "09:00", "duration": "08:00", "room": "Raum B", "slug": "open-source-ai-workshops-2026-88790-ai-agents-in-action-building-smart-open-source-llm-workflows", "url": "https://pretalx.com/open-source-ai-workshops-2026/talk/HY8GHW/", "title": "AI Agents in Action: Building Smart, Open-Source LLM Workflows", "subtitle": "", "track": null, "type": "Workshop", "language": "de", "abstract": "AI-powered autonomous agents are changing the way we interact with large language models (LLMs). In this hands-on workshop, you will learn how to build and customize AI agents using open-source models and frameworks. We will explore how these agents reason, plan, and execute tasks, and how they can be integrated with external tools.\r\n\r\nThis workshop is designed for developers, researchers, and AI enthusiasts who want to build intelligent LLM applications. No prior experience with AI agents is required\u2014just basic Python knowledge and a curiosity to experiment!\r\n\r\n--- After the workshop you will know... ---\r\n\u2705 What AI agents are and how they interact with LLMs\r\n\u2705 How to build AI-driven workflows using open-source frameworks\r\n\u2705 How to integrate external tools and APIs using MCP\r\n\u2705 How to secure your agents using guardrails\r\n\u2705 How context engineering can be used to reduce hallucinations", "description": "AI agents allow large language models to go beyond simple text generation\u2014they can think, plan, and take action. In this workshop, we will explore how to design LLM-powered agents that automate workflows, process information, and interact with external systems.\r\n\r\nIn this workshop you will be able to choose among many different cloud-based LLMs to gather hands-on experience with agentic AI. You\u2019ll start by building a simple AI agent, then extend it with new capabilities, such as integrating APIs or working with structured data.\r\n\r\n--- Key Topics Covered ---\r\n\u2714 Understanding AI Agents: What they are, how they work, and common frameworks\r\n\u2714 Hands-on Development: Building a simple agent from scratch\r\n\u2714 Expanding Capabilities: Enhancing agents with plugins, APIs, or external tools\r\n\u2714 Optimizing Agent Performance: Debugging, logging, and prompt tuning\r\n\u2714 Guardrails: Securing your agentic workflows\r\n\u2714 Context Engineering: How to reduce hallucinations\r\n\r\n--- Who Should Attend? ---\r\n\ud83d\udc68\u200d\ud83d\udcbb Developers interested in LLM-powered automation\r\n\ud83d\udcca Data scientists & AI researchers exploring autonomous AI workflows\r\n\ud83d\udee0\ufe0f Tech enthusiasts looking for hands-on experience with AI agent frameworks\r\n\r\n--- Prerequisites ---\r\n\u2022 Basic Python programming skills\r\n\u2022 Familiarity with Git & working in a terminal\r\n\u2022 Some knowledge of LLMs (not required but helpful)\r\n\u2022 A laptop with Python 3.11+ installed, or the ability to use a cloud-based environment", "recording_license": "", "do_not_record": false, "persons": [{"code": "PK9RWF", "name": "Luca Rolshoven", "avatar": "https://pretalx.com/media/avatars/PK9RWF_Poc9TNt.webp", "biography": "I am a PhD student doing research in NLP. I am working on an interesting Sinergia project that aims to measure the level of sustainability in call for tender documents in the domain of public procurement. Previously I worked as a Senior Machine Learning Engineer at BSI, where I was focusing on the deployment and ethical use of Large Language Models within the BSI Customer Suite. Regarding my studies, I earned my masters's in Computer Science, specializing in Data Science, from the University of Bern. During my master's program, I contributed to the development of DiBB, a Python library aimed at distributed black-box optimization.\r\n\r\nOutside of work, I keep a close eye on emerging trends in generative AI and human-centered technology. When I'm not exploring the digital realm, you'll find me delving into psychology, decoding escape rooms, and indulging in bicycle rides.", "public_name": "Luca Rolshoven", "guid": "96a93816-b900-56c7-80ac-48cbb20f391d", "url": "https://pretalx.com/open-source-ai-workshops-2026/speaker/PK9RWF/"}], "links": [], "feedback_url": "https://pretalx.com/open-source-ai-workshops-2026/talk/HY8GHW/feedback/", "origin_url": "https://pretalx.com/open-source-ai-workshops-2026/talk/HY8GHW/", "attachments": []}], "Raum C": [{"guid": "d03bee76-5489-534b-8b79-841c368a6d45", "code": "YZ8W7Y", "id": 89854, "logo": null, "date": "2026-05-18T09:00:00+02:00", "start": "09:00", "duration": "08:00", "room": "Raum C", "slug": "open-source-ai-workshops-2026-89854-das-vibe-coding-experiment-vier-teams-sechs-stunden-ein-ziel", "url": "https://pretalx.com/open-source-ai-workshops-2026/talk/YZ8W7Y/", "title": "Das Vibe-Coding-Experiment: Vier Teams, sechs Stunden, ein Ziel", "subtitle": "", "track": null, "type": "Workshop", "language": "de", "abstract": "Kann **ein Team** ohne Vorlage in **6 Stunden** ein **funktionierendes RAG-System** bauen \u2013 nur mit einem Open-Source-Coding-Agenten und einer API-Spezifikation? Wir probieren es aus.", "description": "Vier Teams \u00fcbernehmen je eine Kernkomponente: LLM-Inference-Server, Vektor-Datenbank, Dokument-Prozessor und Chat-Frontend. Jedes Team baut seine Komponente from scratch, ohne fertige Frameworks \u2013 nur mit Basislibraries und opencode-ai als Copilot. Am Ende integrieren wir alles live zu einem funktionierenden Gesamtsystem.\r\n\r\nWir arbeiten durchgehend mit offenen Technologien: opencode-ai als Coding-Agent, Open-Weights-Modelle wie DeepSeek oder GLM-4, und das finale System l\u00e4uft auf einem lokalen Open-Source-LLM.\r\n\r\n**ABLAUF**\r\n\r\nDer Workshop besteht aus vier Sessions:\r\n\r\n1. **Fundament** \u2013 Systemarchitektur verstehen, Einf\u00fchrung in opencode-ai, Teambildung\r\n2. **Verstehen & Planen** \u2013 Teams erarbeiten ihre Komponente, erstellen eine Kurzpr\u00e4sentation\r\n3. **Bauen** \u2013 Implementation der Kernfunktionalit\u00e4t, Zwischenstand\r\n4. **Integrieren** \u2013 Komponenten zusammenf\u00fchren, Live-Demo, Reflexion\r\n\r\n**F\u00dcR WEN**\r\n\r\nF\u00fcr alle, die verstehen wollen, wie KI-Systeme unter der Haube funktionieren, und AI-gest\u00fctztes Programmieren mit Open-Source-Tools ausprobieren m\u00f6chten. Du musst kein erfahrener Entwickler sein \u2013 wichtiger ist die Bereitschaft zu experimentieren.\r\n\r\n**VORAUSSETZUNGEN**\r\n\r\n- Laptop mit Browser\r\n- GitHub-Account\r\n\r\nAlles andere l\u00e4uft im vorkonfigurierten GitHub Codespace.\r\n\r\n**WAS DU MITNIMMST**\r\n\r\n- Praktische Erfahrung mit Vibe Coding\r\n- Verst\u00e4ndnis der Kernkomponenten eines RAG-Systems\r\n- Deinen eigenen Code zum Weiterbauen", "recording_license": "", "do_not_record": false, "persons": [{"code": "GUVNJG", "name": "Joel Barmettler", "avatar": "https://pretalx.com/media/avatars/GUVNJG_889Drcs.webp", "biography": "Senior AI-Architect & MSc. Artificial Intelligence @ University of Zurich", "public_name": "Joel Barmettler", "guid": "d5615fff-cb88-5608-b0d5-be0f0c27dc77", "url": "https://pretalx.com/open-source-ai-workshops-2026/speaker/GUVNJG/"}], "links": [], "feedback_url": "https://pretalx.com/open-source-ai-workshops-2026/talk/YZ8W7Y/feedback/", "origin_url": "https://pretalx.com/open-source-ai-workshops-2026/talk/YZ8W7Y/", "attachments": []}]}}, {"index": 2, "date": "2026-05-19", "day_start": "2026-05-19T04:00:00+02:00", "day_end": "2026-05-20T03:59:00+02:00", "rooms": {"Raum A": [{"guid": "7c0aa386-8016-5286-aab8-7d43defd93e5", "code": "YSWYU8", "id": 88631, "logo": null, "date": "2026-05-19T09:00:00+02:00", "start": "09:00", "duration": "08:00", "room": "Raum A", "slug": "open-source-ai-workshops-2026-88631-robotics-ethical-ai-in-industry-how-to-apply-the-11-element-compliance-framework-hands-on-cross-industry-edition", "url": "https://pretalx.com/open-source-ai-workshops-2026/talk/YSWYU8/", "title": "Robotics & Ethical AI in Industry: How to Apply the 11-Element Compliance Framework (Hands-On, Cross-Industry Edition)", "subtitle": "", "track": null, "type": "Workshop", "language": "de", "abstract": "As robotics and AI systems move rapidly into industrial environments, compliance, safety, and ethical governance have become strategic enablers\u2014not blockers\u2014of innovation.\r\nIn this hands-on workshop, you will learn how to apply the newly developed \u201c11 Key Elements for Achieving AI & Robotics Compliance in 2025 and Beyond\u201d to your own industry context.\r\n\r\nThrough interactive exercises, group work, and case-based simulations, you will explore how organizations can operationalize ethical AI, prepare for emerging regulations, and embed governance as a driver of trust, performance, and resilience.\r\n\r\nThis workshop is ideal for professionals across sectors who want practical tools for adapting AI governance frameworks to industrial, public-sector, and cross-organizational environments.", "description": "You will learn:\r\n\r\n\u2705 How to adapt the framework to your own industry (manufacturing, logistics, healthcare, public sector, mobility, etc.)\r\n\u2705 How governance acts as an innovation catalyst, not just a regulatory requirement\r\n\u2705 How to map real-world use cases to risk classes and compliance expectations\r\n\u2705 How to align stakeholders (technical + legal + business) using a shared model\r\n\u2705 How to evaluate maturity and readiness within your organization\r\n\u2705 Bonus: Creating a quick-start \u201cAI Governance Canvas\u201d you can take home", "recording_license": "", "do_not_record": false, "persons": [{"code": "UUMBCN", "name": "Kateryna Portmann", "avatar": "https://pretalx.com/media/avatars/UUMBCN_ST6Xb9r.webp", "biography": "Kateryna Portmann is a senior expert in ethical AI governance, robotics compliance, and autonomous systems, with extensive experience bridging product development, regulation, and real-world deployment of AI-enabled robotics.\r\n\r\nShe currently works as a Senior Product Manager at ANYbotics, where she is directly involved in the development and deployment of autonomous robotic systems in safety-critical industrial environments. This role grounds her governance work in the realities of engineering constraints, operational risk, and market adoption.\r\n\r\nEthical & Responsible AI Governance\r\n\r\nKateryna designs and implements governance frameworks for safe, transparent, and trustworthy AI, with a focus on embedding governance directly into industrial and enterprise product lifecycles. She is a strong advocate for treating governance as an innovation accelerator, enabling scalable and compliant AI rather than slowing it down.\r\n\r\nRobotics & Autonomous Systems Compliance\r\n\r\nWith hands-on experience in industrial robotics, Kateryna applies AI governance principles to autonomous and physical systems, addressing safety, accountability, and risk classification. She actively aligns robotics deployments with emerging global standards, including the EU AI Act, ISO robotics standards, and functional safety requirements.\r\n\r\nMulti-Stakeholder AI Alignment\r\n\r\nKateryna specializes in aligning engineering, legal, cybersecurity, safety, and business stakeholders around shared governance models for high-impact AI systems. She has led and facilitated cross-organizational collaboration to ensure that compliance, risk management, and innovation goals reinforce\u2014rather than conflict with\u2014each other.\r\n\r\nApplied Industry Experience\r\n\r\nHer work is deeply rooted in real-world robotics and AI deployments, particularly in regulated and safety-critical contexts. She consistently bridges strategic policy and technical execution, translating regulatory expectations into actionable engineering and product decisions.\r\n\r\nFramework Creation & Thought Leadership\r\n\r\nKateryna is the co-developer of the \u201c11 Key Elements for Achieving AI & Robotics Compliance in 2025 and Beyond\u201d, a framework informed by applied industry practice and multistakeholder collaboration. She contributed to discussions at AI House Davos 2025 and advocates for governance models that support innovation, transparency, and sovereign AI ecosystems.\r\n\r\nCommunity & Academic Engagement\r\n\r\nBeyond industry, Kateryna is an active contributor to the broader robotics ecosystem. She is involved with IFR Women in Robotics, supporting diversity, leadership, and knowledge-sharing in the field. She also serves as a Lecturer at HWZ (Hochschule f\u00fcr Wirtschaft Z\u00fcrich), teaching robotics and AI with a strong emphasis on governance, safety, and real-world application.", "public_name": "Kateryna Portmann", "guid": "26ae4732-2e60-5d8c-acfe-243480208a19", "url": "https://pretalx.com/open-source-ai-workshops-2026/speaker/UUMBCN/"}], "links": [], "feedback_url": "https://pretalx.com/open-source-ai-workshops-2026/talk/YSWYU8/feedback/", "origin_url": "https://pretalx.com/open-source-ai-workshops-2026/talk/YSWYU8/", "attachments": []}], "Raum B": [{"guid": "9dfae401-cd95-561d-8034-843ff9872947", "code": "QSLE3B", "id": 88573, "logo": "https://pretalx.com/media/open-source-ai-workshops-2026/submissions/QSLE3B/Model_QIcI5TI.webp", "date": "2026-05-19T09:00:00+02:00", "start": "09:00", "duration": "08:00", "room": "Raum B", "slug": "open-source-ai-workshops-2026-88573-empowering-ai-agents-building-intelligent-ai-agents-via-mcp", "url": "https://pretalx.com/open-source-ai-workshops-2026/talk/QSLE3B/", "title": "Empowering AI Agents: Building Intelligent AI Agents via MCP", "subtitle": "", "track": null, "type": "Workshop", "language": "de", "abstract": "In this workshop, we will introduce the ideas behind Agentic AI systems using a recently introduced protocol called Model Context Protocol or MCP [Please see https://en.wikipedia.org/wiki/Model_Context_Protocol]. Throughout the day of the workshop, we will plugin various tools to the LLM using MCP servers and show you how you can implement it for your own work.\r\n\r\nThis is a full day hands-on technical workshop.\r\n\r\nNOTE:\r\n- The workshop will be held in English. \r\n- We will be using Anthropic Desktop Claude as our LLM [Please see https://claude.com/download]. It would be great if you can download the application of your machines and create an account. Please note that you can use the free version but it will have limits [Please see https://support.claude.com/en/articles/8602283-about-free-claude-usage].\r\n- We also expect that you will have access to tools like Github\r\n- If you do not know how to code, you can just use your favourite LLM to generate code during the workshop.", "description": "The broad topics that will be covered are (but not limited to):\r\n- Introduction to Agentic AI systems: background, Basic concepts\r\n- Introduction to MCP: What is it, What problems is it solving\r\n- How to write an MCP server: Local and Remote\r\n- Implement a use-case using MCP\r\n- Use your own use-case and implement it\r\n\r\nWe (highly) encourage you to bring your own datasets or API access so that you can use them to implement your use-case with an MCP server (using Anthropic Claude). You should, of course, keep in mind regulations and privacy laws that apply to your data/API access.", "recording_license": "", "do_not_record": false, "persons": [{"code": "NHJSQE", "name": "Sid Singh", "avatar": "https://pretalx.com/media/avatars/NHJSQE_8j2Xslq.webp", "biography": "I am an AI researcher at the Berner Fachhochshule. My day-to-day activity consists of training Large Language Models, creating RAG systems and everything related to these topics. I also teach many different courses on artificial intelligence for both technical and non-technical students.", "public_name": "Sid Singh", "guid": "56bb3461-e97d-5ed2-977c-012c379fcede", "url": "https://pretalx.com/open-source-ai-workshops-2026/speaker/NHJSQE/"}], "links": [], "feedback_url": "https://pretalx.com/open-source-ai-workshops-2026/talk/QSLE3B/feedback/", "origin_url": "https://pretalx.com/open-source-ai-workshops-2026/talk/QSLE3B/", "attachments": []}], "Raum C": [{"guid": "28a3a754-79b9-57ab-926f-c9808b110691", "code": "SVYPLV", "id": 88603, "logo": null, "date": "2026-05-19T09:00:00+02:00", "start": "09:00", "duration": "08:00", "room": "Raum C", "slug": "open-source-ai-workshops-2026-88603-designing-thinking-machines-the-future-of-ai-with-graphs-based-ai-agents", "url": "https://pretalx.com/open-source-ai-workshops-2026/talk/SVYPLV/", "title": "Designing Thinking Machines: The Future of AI with Graphs-Based AI Agents", "subtitle": "", "track": null, "type": "Workshop", "language": "de", "abstract": "Ever wished your AI could do more than just answer a question?\r\nWhat if you could teach it how to think, decide, and verify?\r\n\r\nThis workshop introduces one of the newest and most powerful AI architectures: graph-based reasoning systems, where AI agents follow structured decision graphs to retrieve information, reason step by step, and produce reliable, evidence-based answers.\r\n\r\nWe begin from first principles, explaining how large language models work and why they cannot be trusted on their own. You will then build a complete RAG pipeline that connects an open-source LLM to real documents, enabling it to retrieve evidence and generate fact-grounded answers instead of confident guesses.\r\n\r\nFrom there, we move beyond linear pipelines.\r\n\r\nYou will transform your RAG system into a graph-based AI agent: an autonomous assistant whose reasoning is explicit, structured, and controllable. Using a reasoning graph, the AI learns when to retrieve information, how to combine multiple sources, when to verify results, and when to stop. Instead of one-shot responses, your system follows a clear decision flow that mirrors human problem-solving.\r\n\r\nBy the end of the workshop, you will have built a complete, open-source AI assistant that can read documents, retrieve knowledge, reason through a graph, and answer with evidence.\r\nNot a simple chatbot.\r\nA truly intelligent AI agent that follows a path of thought.", "description": "## What you will learn:\r\n\r\n- **Master the Fundamentals** : Learn how Retrieval-Augmented Generation (RAG) and AI agents work together to combine retrieval, reasoning, and generation into accurate, context-aware responses.\r\n\r\n- **Build Your Own AI Agent System**: Follow a step-by-step process to build the full pipeline from document ingestion, chunking, embeddings, retrieval, to LLM integration, and then upgrade it into a graph-based agent that knows when to retrieve, what to search, and how to produce grounded answers.\r\n\r\n- **Gain Hands-On Experience**: Work through practical exercises and real-world examples that show how a graph-based agent can explore different type of documents, gather evidence, verify key points, and answer in a reliable way.\r\n\r\n- **Ensure Data Security**: Learn how to deploy open-source AI systems with a strong focus on data privacy, secure processing, and responsible use of sensitive information.\r\n\r\n- **Bonus: Evaluation & Reliability**: Discover how to test and measure your AI system to ensure it is using the right sources, avoiding hallucinations, and producing reliable, trustworthy answers.\r\n\r\n## Who Should Join\r\n\r\nThis workshop is designed for anyone familiar with Python who wants to go deeper into how the most robust and modern AI systems are built. Perfect for:\r\n \ud83d\udcbb Developers\r\n\ud83d\udccaData Scientists \r\n\ud83d\udee0\ufe0fEngineers \r\n\ud83e\udd16Curious AI Enthusiasts \r\nSome experience with Python programming is required. You don\u2019t need any background in machine learning or artificial intelligence. We\u2019ll build everything together, step by step.\r\n\r\n## What to Bring\r\n\r\nJust bring your laptop and a stable internet connection (Wi-Fi). We\u2019ll take care of the rest.\r\n\r\n## Why This Matters\r\n\r\nMost LLMs are impressive\u2026 but also a little dangerous. They often sound confident even when they are wrong.\r\nIn this workshop, you will build an AI system where RAG provides knowledge, AI agents add judgment and graph-based reasoning allows your AI to search, verify, and think (in the right way) before it answers.", "recording_license": "", "do_not_record": false, "persons": [{"code": "BADDMT", "name": "Ornella Vaccarelli", "avatar": "https://pretalx.com/media/avatars/BADDMT_75rNyCP.webp", "biography": "Ornella Vaccarelli is a Senior Research Scientist at iCoSys and the Lead Scientist at SCAI (Swiss Center for Augmented Intelligence), where she pioneers innovative AI solutions across diverse domains. With expertise ranging from computer vision and computational physics to the latest developments in large language models (LLMs), she bridges cutting-edge research and practical application.\r\n\r\nHer collaborative projects span from fundamental research at EPFL to developing an LLM-RAG system for the Swiss parliament. Ornella\u2019s work not only advances scientific understanding but also transforms how industry and government leverage AI for informed decision-making.\r\n\r\nA prolific researcher, her findings have been published in high-impact journals, and she is a regular speaker at international conferences. Ornella earned her PhD in Computational Physics from Sorbonne University in Paris and holds a Master\u2019s in Theoretical Physics from the University of Bari. Her career exemplifies a commitment to pushing the boundaries of AI while ensuring its responsible and effective integration into real-world applications.", "public_name": "Ornella Vaccarelli", "guid": "f9b05fdb-51ae-5ab6-8472-15166bcb8d8f", "url": "https://pretalx.com/open-source-ai-workshops-2026/speaker/BADDMT/"}], "links": [], "feedback_url": "https://pretalx.com/open-source-ai-workshops-2026/talk/SVYPLV/feedback/", "origin_url": "https://pretalx.com/open-source-ai-workshops-2026/talk/SVYPLV/", "attachments": []}], "Raum D": [{"guid": "ce09d11e-9cb2-5bac-a443-331892b1bd66", "code": "DRDPRE", "id": 89125, "logo": "https://pretalx.com/media/open-source-ai-workshops-2026/submissions/DRDPRE/image_j4oELtU.jpg", "date": "2026-05-19T09:00:00+02:00", "start": "09:00", "duration": "05:00", "room": "Raum D", "slug": "open-source-ai-workshops-2026-89125-einfuhrung-open-source-ai-fur-entscheidungstragende", "url": "https://pretalx.com/open-source-ai-workshops-2026/talk/DRDPRE/", "title": "Einf\u00fchrung Open Source AI f\u00fcr Entscheidungstragende", "subtitle": "", "track": null, "type": "Workshop", "language": "de", "abstract": "Dieser Workshop vermittelt Entscheidungstragenden eine fundierte Einf\u00fchrung in die Entwicklung und den Betrieb von Open Source KI rund um technische, rechtliche und praktische Aspekte. Neben den Vorteilen wie Kosteneffizienz und digitale Souver\u00e4nit\u00e4t werden auch Herausforderungen wie technische H\u00fcrden und juristische Fragen beleuchtet, um F\u00fchrungskr\u00e4ften, IT-Verantwortlichen und Innovationsmanagern fundierte Entscheidungsgrundlagen zu bieten.", "description": "Die rasante Entwicklung k\u00fcnstlicher Intelligenz (KI) er\u00f6ffnet vielf\u00e4ltige M\u00f6glichkeiten f\u00fcr Unternehmen und Beh\u00f6rden. Open Source KI-Modelle wie DeepSeek und \u00fcber 2.7 Millionen weitere Modelle auf Hugging Face spielen dabei eine zentrale Rolle, da sie Innovation, Transparenz und Anpassungsf\u00e4higkeit f\u00f6rdern. Wie funktioniert Open Source KI, wie werden sie konkret angewendet und welche Chancen und Herausforderungen ergeben sich aus der Nutzung von Open Source KI?\r\n\r\nDieser Workshop bietet Entscheidungstragenden eine fundierte Einf\u00fchrung in die Entwicklung und den Betrieb von Open Source KI mit einem technischen \u00dcberblick, Einblick in rechtliche und ethische Aspekte und zahlreichen Beispielen aus der Praxis. Wir beleuchten die Vorteile wie Kosteneffizienz, Innovationsbeschleunigung und Unabh\u00e4ngigkeit von propriet\u00e4ren Anbietern (digitale Souver\u00e4nit\u00e4t), diskutieren aber auch Herausforderungen wie technische H\u00fcrden, wirtschaftliche Fragen und die Notwendigkeit von internem Knowhow. \r\n\r\nDer Workshop richtet sich an F\u00fchrungskr\u00e4fte, IT-Verantwortliche sowie Projekt und Innovationsmanager, die fundierte Entscheidungen zur Nutzung von Open Source KI in ihrer Organisationen treffen m\u00f6chten. Der Hauptdozent ist Prof. Dr. Matthias St\u00fcrmer, Leiter des Instituts Public Sector Transformation der BFH. Als Gastreferierende werden Luca Rolshoven (Informatik-Doktorand) und Veton Matoshi (NLP-Spezialist) Beitr\u00e4ge halten. Das Programm ist wie folgt:\r\n\r\n09:00h\t\u00dcberblick Open Source AI -> Matthias St\u00fcrmer (BFH)\r\nKurze Vorstellungsrunde, Fragen abholen, Definition digitale Souver\u00e4nit\u00e4t und Open Source AI, Einf\u00fchrung KI, LLMs, Pre-Training, Fine-Tuning etc.\r\n10:30h\tPause\r\n11:00h\tNatural Language Processing (NLP) -> Veton Matoshi (BFH)\r\nVerarbeitung von Text in Machine Learning Verfahren, Vektorisierung von W\u00f6rtern und S\u00e4tzen, Transformers, Praktische Beispiele mit Python, spaCy und Jupyter Notebooks\r\n12:30h\tMittagspause\r\n13:30h\tDeep-Dive in Open Source AI -> Luca Rolshove (BFH)\r\nAktuelle Tech-Trends in Open Source AI, Vergleich von Open Source und propriet\u00e4ren AI-Modellen, Benchmarks und Leaderboards\r\n14:30h\tPause\r\n15:00h\tAnwendung von Open Source AI -> Matthias St\u00fcrmer (BFH)\r\nBetrieb von Open Source AI Modellen, Retrieval-Augmented Generation mit Open Source AI Modellen, Agentic AI und das Model Context Protocol (MCP)\r\n17:00h\tAbschluss", "recording_license": "", "do_not_record": false, "persons": [{"code": "AVFPE7", "name": "Matthias St\u00fcrmer", "avatar": "https://pretalx.com/media/avatars/AVFPE7_j8dPwPn.webp", "biography": null, "public_name": "Matthias St\u00fcrmer", "guid": "021861e0-2e7c-5387-8a40-8abcb071e45c", "url": "https://pretalx.com/open-source-ai-workshops-2026/speaker/AVFPE7/"}, {"code": "L9RQGX", "name": "CH Open", "avatar": "https://pretalx.com/media/avatars/L9RQGX_nynj8na.webp", "biography": "tbd", "public_name": "CH Open", "guid": "23a6f39d-da01-5b4f-9edb-a0b32f592325", "url": "https://pretalx.com/open-source-ai-workshops-2026/speaker/L9RQGX/"}], "links": [], "feedback_url": "https://pretalx.com/open-source-ai-workshops-2026/talk/DRDPRE/feedback/", "origin_url": "https://pretalx.com/open-source-ai-workshops-2026/talk/DRDPRE/", "attachments": []}]}}]}}}