Internet infrastructure and developer platform.
Cloudflare's dashboard agent — which now invokes four MCP suites to act on a user's behalf — is measured in Braintrust with an LLM-as-a-judge scoring whole conversations for resolution, broken down by product area. The eval diff runs only when a skill, prompt, or recipe changes and blocks regressions in CI/CD; the same suite benchmarks sub-agents against models of varying capacity, and covers Cloudflare's internal runbook automation and AI code reviewer.
cloudflare.comSources last verified Aug 22, 2026
Braintrust Evaluates the dashboard agent and gates skill changes in CI/CD.
Receipts · 1
“That was something we leaned on Braintrust for.” — braintrust.dev, Aug 2026
Source last verified Aug 22, 2026