What is Calmo?
Calmo is an agent-native SRE platform for engineering teams that want to delegate production operations tasks to DevOps agents. It investigates alerts across code, infrastructure, and telemetry, testing multiple possible explanations against production evidence to identify likely causes. Teams can also use it for incident response, migrations, performance optimization, and postmortem generation, while preserving their existing workflow. Calmo connects with production tools and can be hosted on-premise or configured to use a team’s own AI models.
Calmo Features:
Alert Investigation: Calmo analyzes alerts and develops initial theories using signals from code, infrastructure, and telemetry.
Parallel Hypotheses: It investigates several possible causes at once and checks each against evidence from production systems.
Incident Knowledge: Investigations become searchable knowledge that teammates can consult for answers about systems, processes, and infrastructure.
Playbook Support: Teams can upload playbooks and standard operating procedures for AI-assisted investigations, including database debugging and incident response.
Remediation Plans: Calmo can generate remediation plans with code snippets and pull requests for production issues.
Production Integrations: Connections include Datadog, Grafana, GitHub, Kubernetes, PagerDuty, Sentry, Slack, and other production tools.
Deployment Controls: Teams can use their own AI models or host Calmo entirely within their own infrastructure.
Calmo Benefits:
Less Alert Noise: By analyzing alerts and surfacing relevant signals, Calmo helps engineers spend more attention on issues that need action.
Faster Diagnosis: Checking competing explanations in parallel can reduce the time teams spend pursuing a single unconfirmed theory.
Shared Expertise: Searchable investigation records make operational knowledge easier to find without repeatedly interrupting senior engineers.
Broader Team Coverage: Access to investigation knowledge and procedures can help more engineers handle production problems.
Greater Data Control: Model choice and on-premise hosting give teams options for keeping control over AI inference and infrastructure.
Use Cases:
Alert Triage: Use Calmo to examine a new production alert and gather relevant signals before an engineer begins investigating.
Database Debugging: Apply an uploaded step-by-step procedure when investigating a database issue affecting a production service.
Security Incidents: Use a security investigation and containment procedure to guide work on a suspected breach.
Deployment Checks: Use pre- and post-deployment verification procedures when preparing a release or checking its production impact.
API Health: Apply API health monitoring and alerting procedures when investigating problems with an API in production.