The AI SRE for complex production environments

Cleric investigates production issues, captures what works, and makes your whole team faster.
Dashboard screen showing investigation of PagerDuty alert for Checkout API 5xx error spike from 0.3% to 12.4% after deploy v1.4.5 with root cause and evidence graph.Chat thread showing Pagerduty reporting 5xx errors spiked to 12%, Cleric app reporting error rate jump after v1.4.5 deployment with a graph, and a user named Michael requesting a fix.Slack thread under #issues with Pagerduty reporting 12% 5xx errors spike, Cleric explaining deploy v1.4.5 caused blocking sync call exhausting payment service pool, and Michael asking Cleric to create a fix.
5 min
Time to Root Cause
92%
Actionable Findings
200,000+
Production-Grade Investigations
Maxime Fouilleul Image
“I see Cleric functioning as an SRE companion for our Software Engineers. If one team has already solved an alert, we can apply that knowledge elsewhere.”
Maxime Fouilleul Image
Maxime Fouilleul
Head of Infrastructure & Operations | BlaBlaCar
How it works

Meet Cleric

Cleric investigates problems, surfaces answers, and learns from your engineers.
Diagram showing a software platform workflow with sections for Alert, Issue Tracker, User Query on left, central platform featuring Learn, Discover, Investigate steps, and Root Cause, Resolution, PR sections on the right, supported by various technology logos including Google Cloud, Kubernetes, AWS, Azure, and others.Diagram showing a platform that helps learn, discover, and investigate alerts, issue trackers, and user queries to find root cause, resolution, and PR, integrating with production systems, observability, knowledge, and code tools.
Cleric in action

Things break. Cleric investigates.

While you’re still reading the alert, Cleric’s already three steps into the investigation.

Skip straight
to the answer

Skip straight to the answer

Cleric automatically investigates alerts and delivers root cause analysis so you know exactly what's wrong and whether to act.
Slack message from Cleric app showing investigation completed in 2.4 mins with a line graph, indicating high error rates in checkout service caused by Kubernetes and another dependency, plus Create PR and View Evidence buttons.
Investigation detail flowchart showing Issue #3456 with high latency on flight-query-api, detailing causes like faulty deployment, query engine slowdown, inefficient SQL query, and root cause as missing composite index.

Trust but Verify

See exactly how Cleric reached its conclusions so you can trust the diagnosis and move to resolution.

Ship a fix

Work with Cleric to ship fixes by asking questions, giving instructions, and collaborating until the issue is resolved.
Works inside your
Slack thread in #issues channel showing Cleric app asking to create a PR for async call, Michael agreeing with a timeout suggestion, and Cleric confirming PR #4321 is ready with options to open in GitHub or merge.
Chat thread showing Sarah requesting Cleric to merge payment async, followed by Cleric confirming merge of PR #4321 and a message that Sarah has contributed 340 times.

Learns from failure

Engineers leave, context doesn’t. Every resolution becomes institutional knowledge your whole team can use.
Under the hood

Simple on the surface. Serious underneath.

Three purpose-built systems that turn alerts into answers—and make every investigation faster than the last.
Maps your world
Services, dependencies, and ownership mapped automatically. When something breaks, Cleric knows where to look.
Tests every theory
Correlates signals, tests hypotheses in parallel, tracks confidence. Systematic elimination, not guesswork.
Compounds over time
Every fix teaches Cleric something new. Patterns and context accumulate automatically.
Explore Product
Security-first

Paranoid by design

Security Overview panel showing encryption in transit and at rest, audit logs enabled, and active statuses for SSO, SAML, and RBAC, next to a shield icon with a star-like logo.
Integrations

Built to plug right into your stack

Connect your whole stack in an afternoon. Read access by default, write access when you’re ready.

Unblock your engineers. Start in minutes.

Setup takes minutes. Read-only by default.