Remote job
DevOps Engineer
Job details
About this role
Role overview Infrastructure-focused engineering role responsible for the platform that powers an AI media generation product. The scope spans a multi-cloud GPU fleet, warm pools of ComfyUI worker processes, Kubernetes orchestration, and Python-based FastAPI services that tie everything together.
Responsibilities
- Operate and evolve a multi-cloud GPU fleet tuned for generative media workloads - Maintain warm ComfyUI worker capacity so generation requests stay responsive - Build and run Kubernetes-based services with a focus on reliability and scaling - Develop and support Python FastAPI services that handle routing, queuing, and integration - Keep infrastructure cost, latency, and uptime within target ranges as demand grows
Requirements
- Production experience with Kubernetes in cloud environments - Strong Python skills, ideally including FastAPI or comparable web frameworks - Familiarity with GPU-backed workloads and the operational quirks they introduce - Working knowledge of ComfyUI or equivalent generative workflow runtimes - Comfort owning infrastructure end to end in a remote setting
Benefits and work setup
- Remote-first arrangement - Equity included in compensation - Direct collaboration with the engineering team building the AI media stack