BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//pydata-london-2026//talk//TAWYHU
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-TAWYHU@pretalx.com
DTSTART;TZID=GMT:20260605T132000
DTEND;TZID=GMT:20260605T140500
DESCRIPTION:LLMs are increasingly being used to take actions\, call APIs\, 
 and write code. But giving AI agents the ability to run code opens up a su
 rprisingly tricky question: how much control do you actually hand over?\n\
 nThere's a full continuum here\, from structured tool calling at one end t
 o full computer use at the other\, but most developers don't realise how m
 any interesting options live in between. That gap matters\, because the ex
 tremes both have serious trade-offs: pure tool calling is safe but sequent
 ial and limiting\, while full sandboxes or computer use are powerful but c
 omplex\, slow\, and often a hard sell to enterprise security teams.\n\nThi
 s talk introduces Monty\, a minimal Python interpreter written in Rust\, p
 urpose-built for running AI-generated code safely. Unlike traditional sand
 boxing approaches that start with full access and try to lock things down\
 , Monty starts from zero and requires you to explicitly grant each capabil
 ity — meaning the LLM can only interact with the outside world through f
 unctions you wrote\, control\, and can audit. It's a new paradigm: not AI 
 using your tools\, but AI writing its own programs to coordinate your tool
 s.\n\nIn this talk\, you will learn how to think about the control-capabil
 ity trade-off when building AI agents\, where Monty sits on that spectrum 
 and why\, and how to use it with Pydantic AI to replace sequential tool ca
 lls with expressive Python — complete with a live demo traced through Lo
 gfire.\n\nBasic familiarity with Python and LLM tool use is helpful but no
 t required. No prior knowledge of Rust or sandboxing concepts needed.
DTSTAMP:20260602T223429Z
LOCATION:Grand Hall 1
SUMMARY:Keynote: Samuel Colvin: Pydantic Monty & Logfire: Wild LLMs\, from 
 tool calling to computer use - Samuel Colvin
URL:https://pretalx.com/pydata-london-2026/talk/TAWYHU/
END:VEVENT
END:VCALENDAR