Skip to main content
ResolveAI is AI for prod. It works across your code, infrastructure, telemetry, and knowledge to help engineering teams run production more reliably and efficiently. Engineering teams spend up to 70% of their time running production — responding to alerts, debugging incidents, updating runbooks, and coordinating across teams. ResolveAI is built to take on that toil.

What It Does

ResolveAI helps you:
  • Delegate on-call to agents: Put Resolve first on the escalation policy. Resolve takes the page, investigates it across code, infrastructure, and telemetry, and decides whether to close it or escalate it to the on-call engineer. The on-call engineer is paged only when a person is needed, and starts with Resolve’s findings in hand. See Primary On-Call.
  • Co-work with agents to resolve incidents: Bring Resolve into your incident channels to drive collaborative remediation. Agents investigate in parallel across observability tools, code, and infrastructure, build causal timelines, and propose fixes so your team spends less time in war rooms.
  • Automate your production workflows: Encode your team’s best practices as repeatable, automated workflows. Run health checks, generate reports, and execute multi-step investigations on demand.

How It Works

  • Understands your production: Connects to your production systems, tools, and code in a few clicks. Builds an understanding of how everything works together and learns from your runbooks, wikis, chats, and guidance.
  • Reasons like a seasoned engineer: Plans investigations that pursue multiple hypotheses in parallel, orchestrating specialized agents to collect evidence from across your systems. Builds causal timelines linking code changes, infrastructure events, and telemetry.
  • Operates tools like an expert: Works with any tool without requiring expertise in query languages, PromQL, Kubectl, or code. Coordinates multi-tool investigations and takes actions like updating JIRA tickets, generating PRs, or writing docs.
  • Decides when to page a person: Follows the team’s escalation criteria, plain-language rules for what is worth a page, and shows its reasoning and evidence for every escalation decision.
  • Builds a living model of your production: Continuously captures tribal knowledge from every interaction: chats, incidents, runbooks, and feedback. Adapts to your environment and custom instructions so every investigation gets smarter over time.

Overview diagram of ResolveAI

Human control

People stay in charge of production. Resolve escalates a page whenever it is unsure, proposes mitigation actions for human approval instead of running them, and can be taken off the pager at any time. See Safety and control.

Get started with Primary On-Call

Security

Resolve does not train on customer data. All data is encrypted in transit and at rest. Security controls support SOC 2, GDPR, and HIPAA requirements. See Security for details.

Contact Us

Need help with setting up Resolve for your team? Contact us at help@resolve.ai.