Patterns
Patterns uses Loop to investigate production traces and record recurring problems and trends.
What it does
- Evaluation and observability
Features 3
- Severity levelsA severity level on each pattern helps prioritize findings and filter the list by impact.
- Suggested next stepsEach pattern is a saved finding with supporting evidence, an explanation and a suggested next step.
- Frequency measurementsLoop refreshes evidence and frequency measurements for active patterns, with representative traces and monitor charts.
Sources
- Pricing
-
Official documentation · 14 Oct 2026
"The only costs associated with Patterns while in public preview are from either built-in inference or model provider inference."
- Description
-
Official documentation · 14 Oct 2026
"record recurring problems and trends as patterns"
- What it does
-
Official documentation · 14 Oct 2026
"Find recurring failures and regressions."
- Sold within
-
Official documentation · 14 Oct 2026
"Patterns is a Loop automation"
- Name
-
Official documentation · 14 Oct 2026
"Patterns is in public preview"
- URL
-
Official documentation · 14 Oct 2026
"Patterns is in public preview"
- Deployment
-
Official documentation · 14 Oct 2026
"On Braintrust-hosted deployments, Braintrust provides the sandbox."
Also from Braintrust
-
Braintrust Platform
Stores production agent traces in its own datastore, turns the patterns it finds there into evaluation datasets, and scores every release against them.
-
Loop AI agent
Loop is an AI agent that investigates Braintrust logs, experiments, and datasets.
Open Braintrust in the directory
Something wrong here? Send a correction — quote this product id: braintrust-patterns.