Members-Only

Recent Talks & Demos are for members only

Exclusive feed

You must be an AI Tinkerers active member to view these talks and demos.

Sign in View FAQ

February 05, 2025 · Dubai

Observability for AI Agent in Langchain

Learn how to monitor a LangChain AI agent with verbose logging, StdOut callbacks, and LangSmith metrics to track decisions, latency, and errors.

Overview

Run an AI agent built with Python and LangChain on a local machine which uses two tools and an OpenAI LLM. Starting with an architecture diagram, we’ll cover:

Verbose Logging: Track agent decisions in real-time using initialize_agent(…, verbose=True)
Callback Managers: Monitor intermediate steps using StdOutCallbackHandler()
LangSmith: Analyze metrics like latency and errors.

All code will be shared via GitHub for hands-on practice by audience.

Tech stack