Get new jobs by email
- ...ability to make pragmatic architecture and implementation tradeoffs. Preferred experience includes Harbor, Terminal-Bench, SWE-bench, Inspect, sandboxed or containerized agents, Docker, Kubernetes, Modal, Daytona, distributed job scheduling, concurrency controls, database...
- ...built, how an agent trial is scored, and how results are aggregated and compared. Experience with Harbor, Terminal-Bench, SWE-bench, Inspect, or a similar harness is a strong plus. Judgment about task quality. You can spot ambiguous instructions, weak or gameable...
- ...shift handovers to ensure continuous operational coverage. GPU Node & Hardware Triage First-Line Diagnostics: Execute runbooks to inspect node health using nvidia-smi, NVIDIA Data Center GPU Manager (DCGM), and system event logs (dmesg, journalctl). Out-of-Band Management...