Our LLM judge scored worse than chance until we made it compare
We rank an agent's investigation traces by comparing them in pairs and turning the wins into an Elo score. Grading each trace on its own scored worse than chance. Comparison made the ranking work, and it catches regressions our other evals miss.
How we verify Cleric’s production fixes
A correct diagnosis isn’t the same as a fixed problem. We built a verifier that checks Cleric’s fixes against the best ground truth available: production itself.
You’re renting your institutional knowledge
Most AI agents rent knowledge back through forward-deployed engineers. The ones worth buying compound verified memory in systems you own.
How Cleric uses Tailscale to securely automate software operations
An AI SRE has to access the same private resources a human engineer would. We use Tailscale to get there.
Stop Reviewing Agent Output. Start Reviewing Agent Decisions.
When agents can produce anything, the bottleneck shifts from code to judgment. You manage that by reading the decisions, not the outputs.
Why Your AI SRE Needs Memory
Investigation logic is becoming a commodity. What won’t commoditize is operational memory: the ability to capture and persist engineering judgment.
Hooks Won’t Secure Your AI Agent
AI Agents require strict network controls in order to keep the data they handle secure.
Cleric launches the first self-learning AI SRE
We created the first AI site reliability engineer (SRE) agent that continuously learns from every incident so software engineers can focus on building instead of firefighting.
The Self-Improving AI SRE
We’ve proven that our self-learning AI SRE approach works. Now we’re ready to scale.
Agent Engineering Is Not Software Engineering
Agent startups hire differently. ML fundamentals and domain intuition matter more than years of software engineering.
Cleric Named a Cool Vendor in the 2025 Gartner® Cool Vendors™ in AI for SRE and Observability
We’ve been named a Cool Vendor in the 2025 Gartner Cool Vendors in AI for SRE and Observability report. We see this as validation of our approach to building a self-improving AI SRE.
The Hidden Complexity of Building an AI SRE
Building an AI SRE isn’t just connecting an LLM to dashboards. It means reasoning through hidden dependencies, conflicting signals, and cascading failures in live production systems.
What’s Good for AI Agents is Good for Engineers
Engineers and AI agents work best under the same conditions. They need clean codebases, fast feedback, and clear observability.
9 Months In: How BlaBlaCar and Cleric Are Reimagining Incident Response with AI
BlaBlaCar runs one of the most advanced reliability stacks in the industry, but even strong systems hit limits as scale grows. By integrating Cleric into their alerting workflow, they’re using AI to reduce manual toil and bring faster, more consistent investigations to every team.
Why Your Engineers Are Drowning in Alerts
Modern engineering teams are overwhelmed by on-call duties and alert fatigue, leaving little time for innovation. Cleric acts as an AI SRE teammate that intelligently investigates incidents, helping engineers stay focused by offloading the cognitive burden of production support.
What is an AI SRE?
Engineering teams are deploying AI agents to handle production operations. This deep dive shows how AI SREs work: building system understanding, investigating issues, and driving resolution. You’ll learn their current capabilities and limitations, and how they will change the way engineering teams operate.
Introducing Cleric: The first autonomous AI site reliability engineer
We’re excited to announce that we’re building an autonomous AI SRE, called Cleric, backed by Zetta Venture Partners in a $4.3M seed round. Cleric is an AI teammate designed to autonomously manage, optimize, and heal software infrastructure.
Give your on-call a headstart.
Start for free, or talk to us about a plan built for your team’s scale and security needs.
Speak to an engineer