BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Business Events Deutschland//DE
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VEVENT
UID:1a9e8ab2cb7c6be174404bfb753c4124072a6fe4@business.alle-events.de
DTSTAMP:20260921T160000Z
DTSTART:20260921T160000Z
DTEND:20260921T170000Z
SUMMARY:Closing the Continuous Learning Loop: From Agent Traces to Agent 
 Training
DESCRIPTION:AI Performance Engineering Meetup (Munich2) **Zoom link**: [h
 ttps://us02web.zoom.us/j/82308186562](https://us02web.zoom.us/j/82308186
 562)  **Talk #0: Introductions and Meetup Updates** by Chris Fregly and 
 Antje Barth  **Talk #1: Closing the Continuous Learning Loop: From Agent
  Traces to Agent Training** by Rustem Feyzkhanov\, Engineering Manager @
  Snorkel.ai Production traces show where agents fail\, but how do we tur
 n those failures into learning? This talk explores a practical loop from
  traces to reproducible simulations\, verifiable rewards\, and model imp
 rovement through fine-tuning or reinforcement learning. We’ll also look 
 at whenit’s better to change prompts\, tools\, or the agent harness inst
 ead ofupdating model weights.  **Zoom link**: [https://us02web.zoom.us/j
 /82308186562](https://us02web.zoom.us/j/82308186562)  **Related Links** 
 Github Repo: [http://github.com/cfregly/ai-performance-engineering/](htt
 p://github.com/cfregly/ai-performance-engineering/) O'Reilly Book: [http
 s://www.amazon.com/Systems-Performance-Engineering-Optimizing-Algorithms
 /dp/B0F47689K8/](https://www.amazon.com/Systems-Performance-Engineering-
 Optimizing-Algorithms/dp/B0F47689K8/) YouTube: [https://www.youtube.com/
 @AIPerformanceEngineering](https://www.youtube.com/@AIPerformanceEnginee
 ring) Generative AI Free Course on DeepLearning.ai: [https://bit.ly/gllm
 ](https://bit.ly/gllm)\n\nQuelle: https://www.meetup.com/ai-performance-
 engineering-meetup-munich/events/315842409/
LOCATION:Deutschland\, Deutschland
URL:https://www.meetup.com/ai-performance-engineering-meetup-munich/event
 s/315842409/
END:VEVENT
END:VCALENDAR
