Senior AI Compiler Engineer, MLIR

NVIDIA•Austin, WA

About The Position

NVIDIA is seeking a Senior AI Compiler Engineer to join a team building an MLIR-based AI compiler. This compiler will power NVIDIA’s inference engine end to end, with a strong focus on achieving high performance, fast build times, and low memory usage. The role involves ensuring Ahead-of-Time and Just-in-Time usability across both data center and edge environments. This position contributes to NVIDIA's mission as an AI computing company, leveraging GPUs to drive advancements in deep learning, including LLMs, generative AI, recommendation systems, vision, and speech.

Requirements

Bachelor's, Master's, or Ph.D. in Computer Science, Computer Engineering, a related field, or equivalent experience
3+ years of relevant work or research experience in performance analysis and compiler optimizations
Experience with compiler technologies such as MLIR, XLA, and LLVM
Excellent C/C++ and Python programming and software design skills, including debugging, performance analysis, and testing
Ability to work independently, define project goals and scope, and lead your own development efforts
Strong interpersonal skills and the ability to thrive in a fast-moving, dynamic, product-oriented team

Nice To Haves

Understanding of deep learning models, algorithms, and frameworks such as PyTorch and JAX
Experience with GPU kernel generation targeting high performance and fast build times
Proficiency in GPU architecture with CUDA or OpenCL programming experience
A track record of mentoring early career engineers and interns is a bonus

Responsibilities

Develop MLIR-based graph representations and optimizations for future GPU architectures
Partner with framework and hardware teams to enable new model patterns and upcoming GPU architectural features
Define APIs and MLIR dialects
Conduct performance optimizations and analysis
Implement compiler optimizations and kernel generation for neural networks
Contribute to other general software engineering work