BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//pydata-london-2026//talk//KDWRYR
BEGIN:VTIMEZONE
TZID:GMT
BEGIN:STANDARD
DTSTART:20001029T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:GMT
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000326T020000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:BST
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-pydata-london-2026-KDWRYR@pretalx.com
DTSTART;TZID=GMT:20260607T161500
DTEND;TZID=GMT:20260607T170000
DESCRIPTION:AI coding agents are changing how data professionals work. But 
 an AI agent chat session is a stream\, a long conversation that scrolls on
  and on. A good notebook is something different: a sequence of distinct\, 
 well-structured transformations\, each with an explanation and a visible r
 esult. How do you get from the chat stream to that? And how do you see the
  visualizations\, the tables\, charts\, and diffs that make data work legi
 ble?\n\nWe'll trace the historical reasons why the programming notebook st
 yle developed\, what problems it solves\, and what problems it creates. No
 tebooks intermingle three valuable concepts: a live execution environment\
 , a long-running process that caches state in memory\, and a narrative log
  of exploration steps. The long-running process is the key. It's why data 
 scientists use notebooks instead of Python scripts. But this coupling is a
 lso why notebooks are fragile\, unreproducible\, and impossible to product
 ionize. And the kernel's implicit mutable state is a poor fit for AI agent
 s. Unlike databases (explicit state\, declarative interface\, introspectab
 le)\, a notebook kernel degrades as implicit state accumulates across cell
 s.\n\nThis talk introduces the Deconstructed Notebook: a system that gives
  AI-agent-driven data work the structure and visualization of a notebook w
 ithout the notebook's baggage. Claude writes the instructions in the termi
 nal. The PyData Arrow stack\, driven by Ibis and xorq\, handles the comput
 e. A browser companion renders tables\, charts\, diffs\, and lineage live 
 as the work iterates\, organized into distinct steps\, not a scrolling cha
 t log. The key architectural insight is that automatic caching of expressi
 on results to disk replaces the notebook kernel's in-memory state\, lettin
 g each step execute as a self-contained script while preserving the intera
 ctive\, incremental workflow data scientists depend on. The system is buil
 t on xorq\, an open-source library built on Ibis and Apache Arrow\, but th
 e design principles generalize. We'll demo the full workflow live and shar
 e what we learned about building post-notebook tooling for the age of AI a
 gents.
DTSTAMP:20260602T225356Z
LOCATION:Grand Hall 1
SUMMARY:The Future of Notebooks in a Claude Code World** - Paddy Mullen
URL:https://pretalx.com/pydata-london-2026/talk/KDWRYR/
END:VEVENT
END:VCALENDAR