In this hybrid role, you will report to a Hardware Engineering Manager
You will:
- Create architectural models for exploration, development and verification of machine learning accelerators
- Analyze workloads, define and derive representative microbenchmarks for functional & performance goals
- Drive equivalence between actual hardware and simulator implementations, both for correctness and performance correlation
- Build scalable tools for modeling, performance analysis and optimization
- Interact with cross-functional engineering teams to identify opportunities and requirements
You have:
- BS degree in Computer Science or Computer Engineering or equivalent, or equivalent practical experience
- 3+ years on modeling complex, high performance architectures in the industry for functionality and/or performance
- Experience with cycle-level simulation frameworks (SystemC, Gem5 or similar)
- Strong C++ programming skills and familiarity with software development methodologies
- Solid foundation in computer architecture and principles of digital logic design
We prefer:
- 1+ years experience modeling machine learning architectures in C++ simulators
- Track record of building tools used by software and hardware teams for performance understanding and/or debug
- Experience writing and debugging microbenchmarks for performance correlation and correctness
- Strong English communication skills (written and verbal)
