BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//events//Events//EN
CALSCALE:GREGORIAN
X-WR-CALNAME:Closing the Continuous Learning Loop: From Agent Traces to Age
 nt Training
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-PUBLISHED-TTL:PT1H
BEGIN:VEVENT
UID:event-mxvnqvu8@ontown.app
DTSTAMP:20261008T203637Z
DTSTART:20261019T160000Z
SUMMARY:Closing the Continuous Learning Loop: From Agent Traces to Agent Tr
 aining
LOCATION:San Francisco\, United States
DESCRIPTION:AI Performance Engineering Meetup (San Francisco\, Global)\nZoo
 m link: https://us02web.zoom.us/j/82308186562\n\nTalk #0: Introductions an
 d Meetup Updates\nby Chris Fregly and Antje Barth\n\nTalk #1: Closing the 
 Continuous Learning Loop: From Agent Traces to Agent Training\nby Rustem F
 eyzkhanov\, Engineering Manager @ Snorkel.ai\nProduction traces show where
  agents fail\, but how do we turn those failures into learning? This talk 
 explores a practical loop from traces to reproducible simulations\, verifi
 able rewards\, and model improvement through fine-tuning or reinforcement 
 learning. We’ll also look at when it’s better to change prompts\, tool
 s\, or the agent harness instead of updating model weights.\n\nZoom link: 
 https://us02web.zoom.us/j/82308186562\n\nRelated Links\nGithub Repo: http:
 //github.com/cfregly/ai-performance-engineering/\nO'Reilly Book: https://w
 ww.amazon.com/Systems-Performance-Engineering-Optimizing-Algorithms/dp/B0F
 47689K8/\nYouTube: https://www.youtube.com/@AIPerformanceEngineering\nGene
 rative AI Free Course on DeepLearning.ai: https://bit.ly/gllm
URL:https://ontown.app/e/mxvnqvu8-closing-the-continuous-learning-loop-from
 -agent-traces-to-agent-training/
END:VEVENT
END:VCALENDAR