PyData Chennai: LLMs in Production

Move past the demo. This workshop covers what it actually takes to run LLM features in production.

Agenda

• RAG pipelines that hold up under real queries
• Building eval harnesses you can trust
• Latency and cost: caching, batching, and model routing
• Observability for non-deterministic systems

Prerequisites

Bring a laptop with Python 3.11+. Working knowledge of Python required. Ticket includes lunch and compute credits.