onsite
Senior Power and Performance Architect, Machine Learning Hardware - Amazon
ML Engineer
Lead power and performance architecture for custom ML silicon, defining low‑power strategies, optimizing ASIC designs, and collaborating with hardware and software teams to deliver high‑efficiency accelerators.
About the role
Key Responsibilities
- Define and drive power‑aware architecture strategies for machine‑learning accelerators and custom ASICs.
- Develop low‑power design methodologies, perform power budgeting, and create optimization models across silicon, firmware, and software stacks.
- Collaborate with circuit, RTL, verification, and software teams to ensure performance targets are met while maintaining power efficiency.
- Lead hardware verification efforts, including creating test plans, running simulations, and analyzing silicon bring‑up results.
- Mentor junior engineers and influence cross‑functional road‑maps for next‑generation ML hardware.
Requirements
- 10+ years of experience in power architecture, ASIC/SoC design, or related hardware engineering roles.
- Strong expertise in low‑power design techniques, power analysis tools, and performance optimization for ML workloads.
- Proficiency in Python and C++ for modeling, scripting, and developing verification environments.
- Hands‑on experience with hardware verification flows, RTL simulation, and silicon bring‑up.
- Proven ability to work across multi‑disciplinary teams and deliver complex hardware solutions on schedule.