
Neural Engine Compiler & Modeling Architect, Platform Architecture
- Cupertino, CA
- Permanent
- Full-time
- Develop and optimize a compiler for a Neural Engine IP to support efficient execution of state-of-the-art ML workloads
- Define, document and implement C/C++ bit-accurate and transaction level models with SoC and Neural Engine arch teams
- Collaborate with design and verification teams to define C-model interfaces for validation and debug
- Develop and maintain architecture test cases and automated workflows to verify the correct functionality of the models
- BS degree
- Experience coding in C/C++ and Python
- Domain knowledge in at least one hardware IP: ML HW accelerators or processing units such as GPU, image/video, CPUs, or similar
- MS or PhD degree
- 20+ years of experience developing models for hardware validation
- Experience developing compiler for ML accelerators
- Experience developing C++ bit accurate models for hardware verification
- Familiarity with working in a chip development environment with RTL designers and verification engineers
- Experience integrating IP models into chip simulation platforms
- Experience debugging complex models