Braintrust Outage History

31 incidents recorded over the last 365 days, 31 of them resolved. Durations are measured from when PulsAPI first saw the problem to when it cleared, which is usually longer than the vendor's own figure.

Live Braintrust status100.00% uptime history

Every Recorded Braintrust Incident

Braintrust incidents, newest first, with date, duration and status.
DateIncidentDurationStatus
Sep 30, 2026Elevated errors and latency on the US data plane APIThe issue has been resolved.1h 18mResolved
Sep 9, 2026Slow or failed data loading in US regionThe issue causing slow or failed data loading in our US-hosted service has been resolved. We're sorry for the disruption.39 minResolved
Sep 4, 2026Elevated 5XXs on Control Plane APIThe issue causing elevated errors in the Braintrust control plane APIs has been resolved.32 minResolved
Aug 28, 2026Gateway 5xxsThe load spike has subsided and 5xxs are mitigated-104 minResolved
Aug 28, 2026Gateway 5xxsThe deploy was rolled back and gateway 5xxs were mitigated.-149 minResolved
Aug 28, 2026Gateway 5xxsThe faulty deploy was rolled back and 5xxs were mitigated.-1746 minResolved
Aug 19, 2026Upstream package mirror is degraded, Brainstore is not currently affectedThe upstream APT package mirror in AWS us-east-1 has recovered and we're no longer seeing impact when starting new brainstore instances.4h 33mResolved
Aug 3, 2026Gateway 5XXsThe issue has been resolved and service has remained stable following the rollback. We will continue to monitor internally, but this incident is now considered resolved.27 minResolved
Jul 16, 2026High 504 error rate in EU data planeWe have reverted to an older configuration to avoid an AWS outage with Cloudfront.28 minResolved
Jun 12, 2026US Data plane outageA configuration change was identified as the root cause and reverted.-12 minResolved
Jun 4, 2026Issues loading logs pageWe have pushed a fix and the loading issues have been remediated.0 minResolved
Jun 4, 2026Automation degradationWe determined the issue was caused by a role used by the automation service losing required permissions following a configuration change. The issue was reported on May 15 at approximately 1:00 PM PT and resolved on May…-26313 minResolved
May 26, 2026Errors logging into BraintrustOur auth provider has recovered.37 minResolved
May 5, 2026High rate of errors on control planeThe issue was caused by our firewall provider unexpectedly blocking traffic despite safeguards we had in place to prevent this behavior. We are actively working with the vendor on additional mitigation measures to…47 minResolved
Apr 26, 2026Control plane outageProvider has recovered. Error levels back to zero.5 minResolved
Apr 20, 2026Control Plane Database Connectivity IssuesWe're no longer seeing impact from the connectivity issues and we are still investigating the root cause.15 minResolved
Apr 13, 2026Control plane database degradationPerformance quickly returned to normal when the problematic query was identified and remediated, and we are implementing safeguards to prevent similar future issues.-30 minResolved
Mar 31, 2026Elevated control plane errorsError rates have dropped and functionality is restored.12 minResolved
Mar 20, 2026Database load issues with control planeLoad issues have subsided1h 15mResolved
Mar 20, 2026Authorization service in degraded stateThe authorization provider has completed mitigation work and is reporting a full recovery in their systems. We are no longer observing impact to users.31 minResolved
Mar 10, 2026Control Plane ErrorsProvider recovered-14 minResolved
Mar 10, 2026Authorization service in degraded stateThe authorization provider has completed mitigation work and is reporting a full recovery in their systems. We are no longer observing impact to users.19 minResolved
Mar 3, 2026S3 Export Automations failingAffected a small subset of users.33 minResolved
Feb 19, 2026Authorization service in degraded stateWe have not observed any recurrence of issues and our authorization provider has reported full recovery and stable performance across their entire system.2h 1mResolved
Feb 18, 2026Control Plane UI HangingThe issue has been resolved and you should be seeing no slowness in the Braintrust Control Plane UI. Please let us know if you have any issues.3 minResolved
Feb 15, 2026High rate of errors on control planeThe root cause appeared to be an issue with an upstream dependency (database connection pooler), which we have replaced. We will be writing a full post-mortem with more details.1h 42mResolved
Feb 10, 2026Logs Page ErrorsPushed out a fix3h 53mResolved
Feb 4, 2026Traces Appearing to Load IndefinitelyWe resolved the issue by reverting the inciting commit.2 minResolved
Feb 3, 2026High error rate in control planeError levels have returned to normal.-363 minResolved
Dec 17, 2025Elevated BTQL errorsQuery error rates have stabilized.42 minResolved
Nov 24, 2025Degraded data plane performanceWe have not observed any errors since the fix was applied.22 minResolved

Hear about the next one while it is open

Everything above is a record written after the fact. We can email you when the next Braintrust incident opens instead, at the point where knowing still changes what you do.

Free, no password, no card. Unsubscribe in one click.

How to Read This Braintrust Incident Log

Each row is an incident PulsAPI observed, not a summary written afterwards. The duration is wall-clock time between the first failing check and the first clean one, so it includes the window before Braintrust acknowledged anything. Vendor post-mortems typically measure from acknowledgement, which is why their numbers are usually shorter.

Incidents still open have no duration yet and are listed as ongoing rather than being given a running total. The archive covers the last 365 days; anything older has aged out of the window rather than never having happened.

The longest single Braintrust outage in this window ran 4h 33m, against a mean recovery of 56 min. If you depend on Braintrust in a customer-facing path, the longest figure is the one to design around. The mean is what happens on a normal bad day; the maximum is what happens on the worst one.