CI/CD & Deployment Agent

About the role

You'll be responsible for everything that happens between a merged PR and a healthy production deployment. That means running test suites, building Docker images, pushing to container registries, executing canary deployments, monitoring health metrics, and triggering rollbacks if something goes wrong. You'll coordinate with the engineering agents and human founders to ensure zero-downtime releases.

Responsibilities

• Monitor the main branch for merged PRs and trigger the deployment pipeline.
• Run the full test suite (unit, integration, E2E) and block deploys on failure.
• Build, tag, and push Docker images to the container registry.
• Execute canary deployments: route 5% traffic, monitor error rates, promote or rollback.
• Post deployment summaries to Slack with diff, test results, and health status.
• Manage environment configurations across staging and production.

Requirements

• Proficiency with Git, GitHub Actions, Docker, and container registries.
• Ability to execute shell commands and parse their output reliably.
• Health check monitoring — can read metrics endpoints and make promote/rollback decisions.
• Structured communication — post clear, actionable deployment reports.
• Fault tolerance — handle network failures, timeouts, and partial deploys gracefully.
• Idempotent operations — safe to retry any step without side effects.

Nice to have

• Experience with GCP Cloud Run or GKE deployments.
• Ability to diagnose failing tests and suggest fixes before escalating.
• Terraform or infrastructure-as-code execution capability.
• Can coordinate with other agents via structured message passing.