Quickstarts
Observability
Integrate with model and cloud providers to send logs to Braintrust
Evaluation
Get started running evals in Braintrust using experiments
Guides
Playgrounds
Prototype in a workspace designed for rapid iteration and evaluation
Experiments
Run evals on your production data to determine what works and what needs changing
Logs
Send application traces to Braintrust for viewing, filtering, and scoring
Datasets
Create datasets from log data to run evals and iterate based on results
Monitor
Design custom dashboards to better understand your application’s performance
Review
Add human review to your evaluation process