Software Engineer, Frontier Systems - Power Management
Core
Develop and implement system-level software solutions to optimize power usage, monitor consumption patterns, and stabilize grid-level power for large-scale supercomputers used in frontier model training.
Role type
Senior IC software engineer (power management & distributed systems)
Builds
Automated monitoring, detection, and remediation tools for supercomputing infrastructure
Domain
Data center infrastructure / High-performance computing / Power systems
Deliverable
production ML models | infrastructure
Required skills
Python, distributed systems, system-level investigations, automation/scripting, SQL, PromQL, Pandas, electrical engineering concepts (digital signal processing, power systems, FFTs)
Preferred skills
Power characteristics of synchronous workloads, IT hardware design power control, control system fundamentals
Technologies
Python, SQL, PromQL, Pandas, shell scripting
Responsibilities
Develop software to optimize power usage in large-scale supercomputers; Build automation to monitor power consumption and stabilize fluctuations; Design tools for real-time monitoring and remediation of power-related faults; Collaborate with hardware teams to integrate power control requirements into IT hardware design; Drive development of power throttling mechanisms based on workload demands.
Seniority
Senior, hands-on IC