Taming LLMs with Pydantic Parsing, Validating, and Guarding output — Manish Sinha (PyBay 2025)

LLMs can be unpredictable, fragile, and tricky to scale responsibly. In this talk, you’ll learn how to use Pydantic to transform that chaos into clean, typed, validated Python objects you can safely rely on in production.Based on our own real world experience with it, we'll explore how to define schemas, coerce LLM output into those schemas, and add guardrails through validation and structured prompting. For more information, see https://pybay.org/ Follow us on LinkedIn   / pybay