Product / UX / Telemetry & observability

Braintrust

Enterprise AI product development platform with logging, tracing, evaluation, and prompt playground. Focuses on the eval-to-production workflow, iterate on prompts, evaluate with custom scorers, deploy with confidence, monitor in production. Strong enterprise positioning with SOC 2 compliance. Funded by a16z.

Braintrust AI eval/observability platform (braintrust.dev); SDK v3.16.0 (Jun 3, 2026, Apache-2.0) confirmed live; platform backend (Brainstore) proprietary SaaS with hybrid enterprise deploy.

Openness

1 high confidence

1.0

source: closed
license: Apache-2.0(OSI, client SDKs only, Python/TS/Go/Ruby/C#)
platform: proprietary(Brainstore custom DB, evals/tracing/automation all SaaS)
self-host: enterprise-only(hybrid)

Open SDKs but a closed SaaS core (Brainstore backend, dashboards, automation are proprietary); no OSS core, so 'closed' (not open_core) like LangSmith/Galileo per recipe's closed examples. The Apache-2.0 client SDK is instrumentation only and does not open the product.

https://github.com/braintrustdata/braintrust-sdk recorded 2026-06-04
Apache-2.0 license on client SDK; platform itself at braintrust.dev
https://www.braintrust.dev/ recorded 2026-06-04
proprietary backend (Brainstore custom DB), SaaS + hybrid enterprise deployment

Adoption

4 high confidence

4.0

~6.08M PyPI downloads in the last month for the `braintrust` package (primary: pypistats); named production users incl. Vercel, Notion, Coursera, Dropbox, Replit. 1-10M monthly-download band; strongest adoption of the closed-platform set here.

https://pypistats.org/packages/braintrust recorded 2026-06-04
6,080,857 downloads last month for the braintrust package
https://www.braintrust.dev/ recorded 2026-06-04
named customers Vercel, Notion, Coursera, Dropbox, Replit

Capability

5 high confidence

5.0

Deepest eval+automation feature surface in this batch (CI-style quality gates, Topics auto-clustering, continuous scoring), a frontier-definer for eval-first LLM observability; rated 5 above the MLflow/Langfuse C4 anchors on feature breadth.

https://www.braintrust.dev/ recorded 2026-06-04
tracing, evals (LLM/code/human), automation (Topics, quality gates), prompt mgmt, OTel/MCP integrations

Unchanged since 2026-07-30 (last edited, not re-checked)