Build and operate infrastructure for evaluating coding agents and language models. Own end-to-end features across data models, APIs, background workers, and user interfaces; run sandboxed agent trials; develop scoring and task-quality systems; integrate models and benchmarks; optimize performance, concurrency, migrations, and reliability; and review code. The role requires independent ownership of production backend systems and hands-on LLM evaluation experience.

Own and improve multi-cloud production infrastructure across GCP, AWS, and Azure. Build developer platforms, CI/CD pipelines, deployment automation, Terraform modules, and production Kubernetes environments. Lead incident response, root-cause remediation, observability, networking, security, cost optimization, and reliability initiatives. The role requires hands-on automation and programming, collaboration with security and application teams, mentoring, global on-call participation, and regular overlap with U.S. Pacific Time.