Understands hardware-level programming interfaces and models, and designs low-level programming interfaces and models.
Designs and builds low-level programming interfaces with well-defined syntax and semantics.
Develops and implements multi-layered low-level runtimes running in RNGD, kernel level, and user level.
Integrates with open-source deep learning frameworks using low-level programming interfaces and runtimes.
Bachelor’s degree in Computer Science or equivalent work experience.
Strong communication skills for requirement gathering and clarification.
3+ years of experience with strong programming skills in Rust, Python, Golang, or C++.
3+ years of experience in functional programming.
Experience in designing programming interfaces and models.
Experience in designing and implementing low-latency asynchronous execution models.
Experience in kernel-level performance optimizations using cutting-edge technologies.
Experience with PyTorch, including its implementation and execution models across a variety of execution environments (e.g., single chip, multi-chip, multi-server, DP, PP, TP).
