Resources

Learn how to get the most out of Kalmia.

Guides, references, and examples to help you debug and improve your AI agents.

Documentation

Complete guide to setting up Kalmia, instrumenting your agents, and using the dashboard. Covers both the Python/TypeScript SDK and the Braintrust integration.

Read the docs

API Reference

Full reference for Kalmia's REST API — experiment registration, annotation CRUD, trace search, and behavior detection endpoints.

View API reference

Quickstart Guide

Get up and running in under 5 minutes. Instrument your agent with Braintrust, register an experiment, and view your first traces in Kalmia.

Get started

Talk to Us

Have questions about Kalmia, need help with your setup, or want to discuss your use case? Book a call with our team.

Book a call

Guides

Comparing Agent Variants

Set up A/B experiments to compare different prompts, models, or tool configurations across your agent.

Behavior Detection

Define custom behaviors in plain language and let Kalmia's AI automatically tag matching traces.

Understanding Variant Metrics

What each metric means — duration, tokens, turns, tool calls, retries, backtracks — and how to use them to improve your agent.

Using the Kalmia SDK

Instrument your Python or TypeScript agent directly with the Kalmia SDK for zero-dependency tracing.

Example: Register an experiment

After your agent runs are logged to Braintrust, group them into an experiment with one API call.

Request

curl -X POST /api/experiments \
  -H "Content-Type: application/json" \
  -d '{
    "name": "RAG vs no-RAG",
    "correlationIds": [
      "run-abc-123",
      "run-def-456"
    ]
  }'

Response

{
  "id": "exp-uuid-789",
  "url": "https://kalmia.app/dashboard
    ?mode=experiments
    &experiment=exp-uuid-789",
  "traceCount": 2,
  "createdAt": "2026-03-01T..."
}