MatX is looking to develop vertically integrated full-stack solutions from silicon to systems for AGI, requiring expertise in optimizing software for specialized hardware.
Requirements
- Experience optimizing software for specialized hardware, employing techniques such as parallelism, SIMD programming, C, assembly-level optimization, or GPU/CUDA programming
- Language: at least one of assembly, C++, C, Zig, or Rust
- Experience implementing kernels for ML models such as Transformers
- Experience using and implementing distributed parallelism techniques such as AllReduce, AllToAll, data parallelism, tensor parallelism
- Familiarity with how compilers work
Responsibilities
- Design and optimize kernels that interface directly with our hardware
- Work in partnership with our ML Research and Hardware Engineering teams
- Provide expertise and guidance on hardware architecture from a programmer's perspective, ensuring seamless integration with the software stack
Other
- Bachelor of Computer Science or equivalent degree
- Authorized to work in the United States
- Work from our offices in Mountain View Tuesdays-Thursdays
- Commuter stipend available for employees within 60 minutes of the office
- 401(k) with contribution matching
- Health, vision, and dental insurance
- Life and AD&D insurance
