Platform Engineer (Agent Release)
On-siteTaguig, Metro Manila, Philippines
Job Summary
Own the end-to-end CI/CD pipeline for every agent and model change, managing gates from initial commit through sandbox, development, staging, and production. Design automated evaluation score thresholds, security scans, and guardrail tests to prevent manual review, while owning progressive rollouts like canary releases to small traffic slices before full promotion. Build automatic rollback triggers for every stage and manage the shadow-period process to ensure metrics are visible before agents transition to full autonomy. Run model version regression testing to confirm existing agents do not degrade upon new model adoption. Partner with the Agent Runtime Engineer to troubleshoot issues during bad rollouts.
Required Qualifications
- At least 5 years in CI/CD, release engineering, or MLOps
- 1–2 years releasing LLM or agent-based systems
- Has designed and run a progressive rollout strategy (canary, shadow, or blue-green) for something stateful or model-driven
- Automatic rollback triggers built in from the start
- Has gated a release on evaluation scores
- Treating a regression as a hard stop
- Strong Python (or similar) skills
- Solid infrastructure-as-code skills
- Terraform
- CDK
- GitHub Actions
- CodePipeline
- Comfortable building and deploying containerized services with Docker
- Owning their release path through CI/CD end to end
Desired Qualifications
- Direct experience with AWS Bedrock
- Experience with something comparable to AWS Bedrock, including its deployment and versioning quirks
- Made model tiering or cost-routing decisions
- Matched task complexity to model cost instead of defaulting everyone to the priciest option
- Worked somewhere release pipelines needed to produce evidence for an audit
- Can explain the difference between traditional CI/CD (pass or fail) and AI release gating (thresholds and probability)
Hiring someone like this?
Get your role in front of qualified candidates on Sorce.