System Performance Engineer, Consumer Devices

OpenAI OpenAI · AI Frontier · San Francisco, CA · Consumer Products

This role focuses on optimizing the performance of AI-powered consumer devices, specifically profiling, benchmarking, and optimizing performance across embedded hardware, operating systems, and applications. The engineer will define product KPIs, build performance tooling, and drive optimizations from lab to launch.

What you'd actually do

  1. Develop system performance benchmarks, methodologies, and policies to evaluate end-to-end product behavior.
  2. Profile and analyze performance across key product use cases and workloads using custom and industry-standard profiling tools.
  3. Partner closely with engineering teams to identify bottlenecks and drive performance optimizations across the software stack.
  4. Define high-level product KPIs and establish measurement frameworks to measure launch readiness and monitor performance throughout the product lifecycle.
  5. Measure, report, and track metrics including system responsiveness, memory utilization, cold boot performance, long-term stability, and behavior under constrained conditions such as thermal limits and low battery states.

Skills

Required

  • Deep experience shipping embedded or high-performance systems on Linux, Android, or iOS devices.
  • Strong proficiency in C/C++, Python, or Rust, with exceptional operating systems fundamentals, including process management, IPC, drivers, interrupts, power management, media frameworks, graphics, filesystems, and networking.
  • A proven track record shipping real-time systems operating under tight latency, power, and memory constraints.
  • Expertise with profiling and debugging tools such as Perfetto, GDB, LLDB, perf, ftrace, or Instruments.
  • Experience building instrumentation, analytics, and performance tuning frameworks.

Nice to have

  • Familiarity with statistics, experimentation methodologies, or data science concepts.
  • Experience collaborating cross-functionally with embedded software, systems, camera and vision, audio, graphics, applications, and hardware teams.
  • Experience defining product KPIs, benchmarking methodologies, and launch readiness metrics for hardware products.
  • Experience investigating field performance regressions and driving fixes through cross-functional collaboration.

What the JD emphasized

  • tight latency, power, and memory constraints
  • tight latency, power, and memory constraints