BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//pyconde-pydata-2026//speaker//ZUECPC
BEGIN:VTIMEZONE
TZID:CET
BEGIN:STANDARD
DTSTART:20001029T040000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:CET
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000326T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:CEST
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-pyconde-pydata-2026-RRLTBU@pretalx.com
DTSTART;TZID=CET:20260414T114500
DTEND;TZID=CET:20260414T131500
DESCRIPTION:**Code-generating LLMs have matured** to the point where they c
 an reliably scaffold **data pipelines and data agents**\, when used in a *
 *supervised\, engineering-first workflow**. This tutorial demonstrates how
  to combine modern **AI coding assistants** with a **production-ready Pyth
 on deployment platform (Tower.dev)** to build and operate **real data syst
 ems**.\n\nParticipants will learn how to structure **collaborative Human/A
 I Assistant development loops**\, where engineers provide **architecture\,
  domain knowledge\, and review**\, while AI accelerates implementation. We
  will build a **data pipeline** and a **lightweight data agent**\, iterati
 ng with an AI assistant to **generate\, test\, and improve code**.  \n\nTh
 e session also covers critical **operational concerns** such as:\n- **Secu
 rity**\n- **Scaling**\n- **Observability**\n- **Debugging**\n\nYou will al
 so see how **production feedback can be looped back into the assistant** t
 o continuously improve generated code.\n\nThis is **not about “vibe codi
 ng”** a website. It is about **disciplined\, review-driven AI collaborat
 ion** that meaningfully improves productivity for **data practitioners at 
 all levels**.
DTSTAMP:20260412T141726Z
LOCATION:Ferrum [2nd Floor]
SUMMARY:From Prompt to Production: How to use AI Code Assistants for Python
  Data Systems - Serhii Sokolenko
URL:https://pretalx.com/pyconde-pydata-2026/talk/RRLTBU/
END:VEVENT
BEGIN:VEVENT
UID:pretalx-pyconde-pydata-2026-WAJQR7@pretalx.com
DTSTART;TZID=CET:20260414T163000
DTEND;TZID=CET:20260414T173000
DESCRIPTION:Software engineering is changing fast. With AI now writing and 
 reasoning about code\, does it still make sense to learn Python or any lan
 guage at all?\n\nIs this the evolution of our craft\, a true revolution\, 
 or just hype from those who benefit most? Join us to debate the future of 
 Python\, the risks of AI-driven development\, and what skills will actuall
 y matter next.
DTSTAMP:20260412T141726Z
LOCATION:Merck Plenary (Spectrum)  [1st Floor]
SUMMARY:Panel: Evolution\, Revolution\, or Illusion? The Future of Python a
 nd Coding in the Age of AI - Sebastian Neubauer\, Markus Klein\, Asya Meln
 ik\, Serhii Sokolenko
URL:https://pretalx.com/pyconde-pydata-2026/talk/WAJQR7/
END:VEVENT
BEGIN:VEVENT
UID:pretalx-pyconde-pydata-2026-YZM8TA@pretalx.com
DTSTART;TZID=CET:20260416T154500
DTEND;TZID=CET:20260416T161500
DESCRIPTION:The AI world is buzzing with claims about “agentic intelligen
 ce” and autonomous reasoning. Behind the hype\, however\, a quieter shif
 t is taking place: Small Language Models (SLMs) are proving capable of man
 y reasoning tasks once assumed to require massive LLMs. When paired with f
 resh business data from modern lakehouses and accessed through tool callin
 g\, these models can power surprisingly capable agents.\n\nIn this talk\, 
 we cut through the noise around “agents” and examine what actually wor
 ks today. You’ll see how compact models such as Phi-2 or xLAM-2 can reas
 on and invoke tools effectively\, and how to run them on development lapto
 ps or modest clusters for fast iteration. \nBy grounding agents in busines
 s facts stored in Iceberg tables\, hallucinations are reduced\, while Iceb
 erg’s read scalability enables thousands of agents to operate in paralle
 l on a shared source of truth.\nAttendees will leave with a practical unde
 rstanding of data agent architectures\, SLM capabilities\, Iceberg integra
 tion\, and a realistic path to deploying useful data agents - without a GP
 U farm.
DTSTAMP:20260412T141726Z
LOCATION:Platinum [2nd Floor]
SUMMARY:Demystifying Agentic AI Using Small Language Models - Serhii Sokole
 nko
URL:https://pretalx.com/pyconde-pydata-2026/talk/YZM8TA/
END:VEVENT
END:VCALENDAR