FluidStack
Facilities Production Engineer, Power
San Francisco, CA$150k–$204kfulltimemidAdded today
About this role
Fluidstack seeks a Facilities Production Engineer to build the software and analytics layer for managing massive-scale AI data center power infrastructure. You'll develop real-time monitoring, capacity modeling, anomaly detection, and automation across utility-to-busway power chains handling GW-scale loads.
What you'll do
- Design software and analytics for power chain management from utility through EPMS to operational decisions
- Build real-time power capacity and consumption models per site for operations visibility and headroom tracking
- Develop detection systems for power anomalies, imbalance, harmonics, and drift before they cause load failures
- Automate post-event power analysis to reduce review cycles from days to minutes
- Operate and maintain power infrastructure across multiple concurrent construction and live production sites
- Write production-grade code for power monitoring and control systems
What they're looking for
- Software engineering for power systems (EPMS, metering, power quality)
- Electrical distribution and power systems knowledge
- Production code development (not just analysis notebooks)
- Time-series data analysis and monitoring systems
- Power quality analytics
- Real-time monitoring and alerting architecture
- Data center infrastructure fundamentals
- Anomaly detection and statistical analysis
Opens the official application on the employer’s site. No login required.
FluidStack
FluidStack builds AI infrastructure at scale, developing data centers and warehouse operations designed to handle gigawatt-capacity compute deployment. The company is hiring for warehouse engineers, data center operations specialists, product engineers, and people leaders to support rapid infrastructure expansion across multiple sites.
- Website
- fluidstack.io
Likely interview questions
- Describe a time you built monitoring that caught a real power or infrastructure event early—what was the impact and how did you validate it worked?
- Walk us through how you'd architect real-time capacity modeling for a data center pulling multi-GW loads with live operations ongoing during construction.