Product Engineer, Compute Operations
Fluidstack · New York, NY
About this role
## About Fluidstack We exist to make humanity more free. Powerful AI will be the biggest lever for human choice—but only if models are aligned with what humanity actually wants. Fluidstack delivers **10 to 100s of GW of compute faster** than anyone else by acquiring power, designing/building data centers, and operating them end-to-end. ## How We Operate - **Be a barrel:** Full autonomy. Own things end to end. - **Insane urgency:** Move fast. - **Reason from first principles:** Challenge assumptions; best idea wins. - **Love of the game:** High-intensity long hours. - **Build something that actually matters:** Work that improves the world. ## The Decision Team (Examples of Key Problems) - **Automate delivery of gigawatts:** Turn land-to-live compute into software (schedules, decisions, and tasks from a live knowledge graph). - **Forward-deploy beside experts:** Convert on-site judgment into systems that reach every unit. - **Deliver every supercomputer faster:** Learn once, prevent issues everywhere. ## Role Scope — Product Engineer, Compute Operations You’ll help build systems that keep and scale frontier compute infrastructure, including: - **Fleet health system:** Real-time telemetry + tiered health checks across **Kubernetes and bare metal**, exposed via a trusted API (“is this machine healthy?”). Alarms correlate into incidents with drafted probable causes. - **Repair & RMA as generated work:** A tracked flow from failure detection → triage → parts/vendor return → return to service. Failure thresholds route machines to repair automatically; each engineer’s shift todo list is generated; time-to-return is reported. - **Hardware qualification as software:** Burn-in, performance baselining, and new hardware validation as rack-level workflows, with acceptance evidence attached in the graph. - **Facility operations on the same system as the fleet:** Roll out maintenance workflows (lockout/tagout, work orders), migrate asset registers, retire legacy inventory before new building energizes. - **Runbooks as checked procedures:** Turn SOPs, training records, and technician qualifications into structured, auditable data; report site SLOs, deployment cycle time, and labor ramp on dashboards. ## What We’re Looking For - Shipped production code in **Go, Python, or TypeScript** (use whatever language the problem demands). - Built real features on **LLM APIs** (OpenAI, Anthropic, or open-weight), **MCP servers**, and/or **agentic frameworks**. - Work daily with AI coding tools like **Claude Code** and **Cursor**, and leverage agents to do useful work autonomously. - Identify problems, design solutions, and ship without waiting for approval. - Move fast under deadlines while leaving foundations others can extend. - Experience with on-call or working alongside on-call teams; convert operational pain into systems. - Strong product taste: interfaces/workflows match how the work actually happens. ### Bonus - Production engineering or SRE on large **GPU fleets** - Hardware qualification / burn-in frameworks - **BMC/Redfish/IPMI** tooling - **CMMS/DCIM/asset management** systems - **BMS/EPMS/SCADA** - **Prometheus/Grafana** ## Benefits - Competitive total compensation (**cash + equity**) - Health, dental, and vision insurance - Retirement plan - Generous PTO ## Equal Opportunity Fluidstack is an Equal Employment Opportunity Employer. All qualified applicants will receive consideration without regard to protected characteristics. ## Apply https://j
Listing freshness
CronJobs last confirmed this listing 2h ago. If its source stops confirming the opening for seven days, this page is removed from active inventory.