ML Accelerator Performance Validation Engineer, Post Silicon Validation

What you'll do

• Design and run performance benchmarks from microarchitectures through full model training; profile real ML workloads.
• Analyze throughput, bandwidth, latency, and bottlenecks, and collaborate with architecture teams on optimization.
• Automate performance regression tracking and compare silicon measurements with RTL simulation and emulation predictions.

What they ask for

• 3+ years of non-internship professional software development experience and 2+ years of non-internship design or architecture experience.
• ML and LLM fundamentals or experience with PyTorch or JAX.
• Bachelor's degree in computer science, engineering, or mathematics, or equivalent; or experience in Java, C++, Python, or a related language.
• 3+ years using hardware performance counters and profiling tools; understanding of computer architecture and experience applying statistical methods to performance data.

Nice to have

• 3+ years of full software development life cycle experience.
• Experience with CUDA or ML/low-level kernels, production LLM deployment on accelerator hardware, collective communications, or HBM/PCIe/DMA bandwidth characterization.

Pay

143,700.00 - 194,400.00 USD annually (base salary)

Tools

PyTorch, JAX, Java, C++, Python, CUDA, Neuron, TPU

AI focus

Validates and optimizes AI/ML training accelerators using ML workloads, including transformer models, LLMs, and vision models.