AI Agents & Evals: Focused Problem Clinic with Soumendra Kumar Sahoo

Stuck on one specific problem while building with AI? Bring it to this focused 25-minute working session.

Choose one topic: an agent workflow that loops or fails, how to start an evaluation dataset, what to trace in an LLM application, or how to compare tools for a concrete use case.

We will examine what you have tried, work through likely causes or trade-offs, and agree on the next experiment or implementation step. Capture the recommendations as we go.

Best for: builders who have a clear question and need practical direction.

Share a short problem statement and a sanitized example before the call. For an end-to-end system review or several connected issues, choose the 55-minute AI Observability & Architecture Review.