Braintrust shows heavy evals, posts role
Braintrust published a write‑up about evaluation pipelines and posted a Recruiting Coordinator role on March 24, signaling operational scaling while teams run heavy production‑scale evals and transcript suites. (dev.to) (linkedin.com)
The Waxell comparison uses a concrete drop‑in example where a team runs 500 real production transcripts every Friday through Braintrust scorers and still records an 8.7/10 evaluation while missing runtime governance issues. (dev.to) The write‑up details Braintrust’s browser‑based evaluation workflow, its Loop feature that suggests prompt improvements and iterates automatically, dataset versioning, and the flow of production traces back into the eval loop. (dev.to) The post lists Braintrust’s usage/pricing signals cited in the piece: a free tier that includes 1 GB processed data/month and 10k evaluation scores and a Pro plan priced at $249/month covering 5 GB processed data before overages. (dev.to) Braintrust’s careers page describes the Recruiting Coordinator role as an in‑office San Francisco position responsible for owning interview scheduling and logistics, building recruiting systems, guiding candidates through the process, and supporting onboarding, with a baseline requirement of 1+ years’ coordinating or people‑ops experience. (braintrust.dev) That Recruiting Coordinator opening is mirrored across third‑party listings and applicant platforms, appearing on Ashby‑hosted job pages and aggregator sites like Built In and ZipRecruiter, showing the posting’s wide distribution. (jobs.ashbyhq.com) Braintrust presents itself as an “AI observability platform” and names customers including Notion, Stripe, Zapier, Vercel, and Ramp on its public hiring and product pages. (braintrust.dev) A public GitHub repository of Braintrust evaluation examples provides code and walkthroughs for building eval suites and connecting eval outputs to production traces, underscoring how teams can reproduce the platform’s evaluation workflows. (github.com)