Continuously Improving Agents with Langfuse — Annabell Schäfer & Lotte Verheyden, ClickHouse
AI Engineer — Oct 11, 2026 — talk
Why watch
A hands-on workshop demonstrating how to build monitoring and evaluation workflows for AI agents using Langfuse with TypeScript and OpenAI Agents SDK. The presenters show how to create evaluators for user feedback, detect issues like all-caps frustration, and implement scope checks against system prompts. The workshop demonstrates turning production failures into improvement cycles through seeded traces and offline experiments.