DevOps Engineer Remote

This contract will begin with a 1 month paid trial which can then extend to 6+ months

We're hiring experienced Dev Ops engineers to author and validate Infrastructure-as-Code (IaC) tasks. You'll design realistic infrastructure scenarios, build the ground-truth solutions, and define the automated checks that grade whether an AI agent solved them correctly and safely. This is hands‑on engineering and judgment work — you're encoding what "good" looks like for real infrastructure operations into verifiable tasks. If you've spent years writing Terraform, Pulumi, Cloud Formation, cloud CLI for AWS or GCP, we want to work with you.

• Author IaC tasks grounded in real-world AWS and GCP scenarios.
• Build ground-truth solutions for each task: correct, idempotent IaC that converges to the desired end state.
• Design verifiable graders — automated checks that confirm an agent reached the correct end state.
• Review and QA tasks authored by other engineers for correctness, difficulty calibration, and robustness.
• Harden tasks against reward hacking.
• Document task intent, assumptions, edge cases, and scoring rationale clearly.

• 4+ years in Dev Ops/cloud engineering.
• Deep, hands‑on Infrastructure-as-Code expertise:
  Terraform (required);
  Pulumi, Cloud Formation, or CDK a plus.
• Strong AWS and GCP depth.
• Comfortable scripting in Python (and/or Bash) to build and automate validation.
• A clear sense of correctness and verification — you can articulate, in code, what makes an infrastructure outcome right and safe.
• Prior work in AI evaluation, benchmarking, or expert data annotation preferred.
• Prior experience with Local Stack preferred.