BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//talks.osgeo.org//qgis-uc2026//talk//LLQG7R
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-qgis-uc2026-LLQG7R@talks.osgeo.org
DTSTART;TZID=CET:20261005T123000
DTEND;TZID=CET:20261005T130000
DESCRIPTION:AI coding assistants are easy to overhype. In practice\, the in
 teresting question is much narrower: where do agentic coding tools actuall
 y help when building and maintaining a real QGIS plugin?\n\nThis talk pres
 ents lessons from developing qfit\, a QGIS plugin for importing\, analysin
 g\, and publishing personal fitness activity data\, using OpenClaw as an a
 gentic coding environment. Rather than treating AI as a magic code generat
 or\, I will focus on practical workflows: breaking work into small reviewa
 ble changes\, using agents to explore code\, draft implementations\, refac
 tor modules\, and strengthen tests\, while keeping a human firmly responsi
 ble for architecture\, validation\, and release quality.\n\nI will share w
 hat worked well\, where the approach failed or created extra review burden
 \, and which guardrails mattered most\, especially around testing\, CI\, c
 ode review\, and incremental design in a PyQGIS codebase. Attendees will l
 eave with concrete patterns for using agentic coding in QGIS plugin develo
 pment without lowering engineering standards.\n\nThis session is aimed at 
 developers and technical QGIS users who are curious about AI-assisted plug
 in development but want concrete practices rather than hype.
DTSTAMP:20260724T192643Z
LOCATION:Bridge 2
SUMMARY:Building a QGIS plugin with agentic coding: lessons from qfit and O
 penClaw - Emmanuel Belo
URL:https://talks.osgeo.org/qgis-uc2026/talk/LLQG7R/
END:VEVENT
END:VCALENDAR