Braintrust
Braintrust sells an observability and evaluation platform for AI agents. It traces agent calls in production, runs experiments against datasets, turns production traces into eval datasets, and its Loop assistant generates prompts and scorers.
Products
Selected offerings — the AI products recorded here, not the vendor's full catalogue.
-
Stores production agent traces in its own datastore, turns the patterns it finds there into evaluation datasets, and scores every release against them.
Inside Braintrust
-
Loop is an AI agent that investigates Braintrust logs, experiments, and datasets; also fixes code and opens pull requests for review through a connected GitHub repository.
-
Patterns uses Loop to investigate production traces and record recurring problems and trends.
-
Braintrust's production-tracing pillar: every prompt, response, tool call and retrieved context logged from the SDK or its gateway, searchable across millions of logs, with latency, cost and quality scored live and alerted on when they drift; also breaks down LLM spend by model, feature or team.
-
Braintrust's offline-evaluation pillar: versioned datasets built from production logs or curated by hand, scorers written as code, as LLM judges or as human review, side-by-side experiments across prompts and models, and CI gates that block a regression before it ships.
-
Braintrust's investigation pillar: it applies a model to a project's own production traces on a schedule, surfaces recurring failure modes as evidence-backed patterns, classifies traffic into topics, and turns a confirmed finding into a regression dataset and scorer.
Closest alternatives 15
Similar products 93
Closest alternatives are pairs a vendor page itself compares. Similar products are matched from what each product does; nothing in that group is asserted as a rivalry.
Sources
- Stage, from total funding
-
Company disclosure · 17 Feb 2026
$125M is the sum of two figures Braintrust states itself, and the company has published no cumulative total since the Series A. Its Series A post, 2024-10-08: "we've raised $36 million to advance the future of AI software engineering, bringing our total funding to $45 million." Its Series B post, 2026-02-17: "Braintrust has raised $80M to become the observability layer for production AI", led by ICONIQ with Andreessen Horowitz, Greylock, Elad Gil and basecase returning -- that post states no running total. $45M + $80M is arithmetic on the company's own two numbers, not an estimate,.
- Valuation
-
SiliconANGLE (2026-02-17): the $80M Series B brings "Braintrust's valuation to $800 million". Reported by press; Braintrust's own post states no valuation.
- Founded
-
AIM Media House: "Launched in August 2023, Braintrust quickly gained traction." A launch date read as the founding year; the company's own pages state none.
- Description
-
Company disclosure · 18 Sep 2026
Own homepage, read 2026-09-18: "Ship quality agents at scale -- Discover patterns in production, turn them into evals, and improve quality with every release." Its named surfaces are Observe, Evaluate, Discover, the Loop agent, pattern automations and Brainstore. Delivery test: the platform stores and scores the customer's own production agent traces.
Open Braintrust in the directory
Something wrong here? Send a correction — quote this company id: braintrust.