Sr Soc Verification Engineer, Cloud-scale Machine Learning Acceleration - Annapurna Labs

Amazon Amazon · Big Tech · Cupertino, CA · Applied Science

This role is for a Senior SOC Verification Engineer focused on custom silicon for ML acceleration in Amazon's datacenters. The engineer will be responsible for verifying these custom chip designs at the SOC level, integrating third-party IPs, creating testplans, writing testcases, executing coverage plans, and debugging bugs. The role also involves mentoring junior engineers and maintaining regression infrastructure. While the chips are for ML acceleration, the core function of the role is hardware verification engineering, not direct AI/ML model development or research.

What you'd actually do

  1. Verify custom chip designs at the SOC level
  2. Integrate 3rd party IPs and VIPs into the SOC testbench
  3. Create comprehensive testplans, write robust random testcases, and execute coverage plans
  4. Maintain autosmoke and regression infrastructure
  5. Dive deep into bugs and triages

Skills

Required

  • 8+ years of design verification experience using System Verilog and UVM
  • 8+ YOE in testbench development including: stimulus, checkers, assertions and coverage

Nice to have

  • Experience verifying multiple levels of design including: custom blocks, IP blocks, sub-systems, and fullchip SOC system testing.
  • Experience using multiple verification platforms.
  • Experience with C/C++ and Object-Oriented Programming.
  • Experience with scripting languages, for e.g. Python.
  • Experience with AMBA protocols, for e.g. AHB/APB/AXI
  • Experience with interconnect protocols, for e.g. PCIe and UCIe
  • Experience with memory sub-systems, including cache coherency and synchronization techniques
  • Experience with verifying complex CPU, GPU, or ML accelerator designs