LLM Observability with Braintrust

This documentation follows the Diátaxis framework for technical documentation, organized into four categories:

• Tutorials: Learning-oriented guides for beginners
• How-To Guides: Task-oriented instructions for specific problems
• Explanation: Understanding-oriented content about concepts and architecture
• Reference: Information-oriented technical details

Tutorials

Step-by-step guides to get you started:

• Using the sampled_traced Decorator - Learn how to efficiently trace high-volume LLM functions

How-To Guides

Task-oriented guides for specific needs:

• Debug LLM Agents Using Braintrust - Leverage Braintrust traces and monitoring for debugging
• Integrate a New Agent with Braintrust - Step-by-step instructions for adding Braintrust to a new agent
• Add a New Agent to CI/CD Evaluations - Configure GitHub Actions to evaluate a new agent in CI/CD

Explanation

Conceptual information about the architecture and design:

• Braintrust Integration for LLM Agents - Overview of how we've integrated Braintrust into our LLM agents

Related Resources

• Braintrust Documentation - Official Braintrust documentation
• Braintrust Python SDK - Official Python SDK