Intern Performance Analysis Engineer (ML Accelerator)

About hirify.global

hirify.global is building hardware for frontier intelligence, focusing on inference workloads with custom ML accelerators. The company is backed by top-tier investors and staffed by leading engineers.

What you will do

• Build components of performance analysis and profiling infrastructure for ML accelerators.
• Collect and analyze performance data including hardware counters, execution traces, and memory behavior.
• Develop tooling to trace host runtime, system behavior, and accelerator execution.
• Correlate performance events across CPUs, accelerators, storage, networking, and distributed workloads.
• Create analysis and visualization tools to identify performance bottlenecks and optimize models.
• Collaborate with hardware, compiler, firmware, and inference engineers to improve developer productivity.

Requirements

• Location: On-site in San Jose, United States
• Strong programming skills in C++ or Rust; Python is a plus.
• Solid understanding of computer architecture including CPUs, GPUs, AI accelerators, memory hierarchies, and parallel programming.
• Experience or strong interest in low-level performance analysis, profiling, and optimization.
• Familiarity with performance analysis tools such as Nsight, VTune, Xprof, Perfetto is a plus.
• Experience or interest in operating systems, compilers, firmware, drivers, or low-level systems software.
• Strong problem-solving skills and ability to learn quickly in a fast-paced environment.

Nice to have

• Experience developing performance analysis or debugging tools.
• Experience with ML accelerator architectures (GPUs, TPUs, etc.).
• Experience with kernel-mode driver development (Linux or Windows).

Culture & Benefits

• Fully in-person team based in San Jose (Santana Row).
• Strong emphasis on engineering skills and cross-disciplinary collaboration between engineering and research.