
Evals is Pydantic's code-first evaluation framework that uses assertions to continuously assess AI agent output quality. It's used by QA engineers and AI product managers ensuring agent outputs meet quality standards, sitting as a continuous evaluation layer that runs alongside production monitoring.
A full-stack AI observability and governance platform built on OpenTelemetry that provides end-to-end visibility…
View SolutionPydantic AI is a typed Python agent framework from the Pydantic team. It lets developers build AI agents with structured outputs and function tools, and switch between many model providers, falling back to another when one fails. It also supports dependency injection, OpenTelemetry-based instrumentation and durable execution, and open-source projects and organisations build on it.
View Solution