Senior AI Compiler Engineer - DL & GPU Optimizations

Visa Hunt; United States; On-site; USD 152,000 - 288,000; Full time

Job summary

NVIDIA is seeking versatile software engineers for its XLA team. You will develop compiler optimization algorithms for deep learning workloads and optimize inference/training for JAX and OpenXLA on NVIDIA GPUs at scale.

You will collaborate with DL framework teams, work with MLIR/LLVM backends, and contribute to features across GPU hardware and software stacks in a fast-paced, production-focused environment.

Qualifications

• Bachelor's, Master's or Ph.D. in Computer Science, Computer Engineering, or related field (or equivalent experience).
• 4+ years of relevant work or research experience in performance analysis and compiler optimizations.
• Ability to work independently, define project goals and scope, and lead your own development effort with clean software practices.
• Excellent C/C++ programming and software design, debugging, performance analysis and test design.

Responsibilities

• Develop compiler optimization algorithms for deep learning workloads on NVIDIA GPUs at scale.
• Collaborate with deep learning framework teams and hardware architecture teams to accelerate next-gen AI systems.
• Code-generation for NVIDIA GPU backends using MLIR, LLVM, and OpenAI Triton; design user-facing features in JAX and related libraries.
• Work with GPU hardware engineers to design AI compiler software features for future GPUs.

Skills

C/C++

Performance analysis

Compiler optimizations

GPU architectures

CUDA/OpenCL

Distributed programming

Interpersonal skills

Education

Advanced degree in CS/CE (Bachelors/Masters/PhD)

Tools

XLA

TVM

MLIR

LLVM

OpenAI Triton

Job description

NVIDIA is seeking versatile software engineers for its XLA team. You will develop compiler optimization algorithms for deep learning workloads and optimize inference/training for JAX and OpenXLA on NVIDIA GPUs at scale.

You will collaborate with DL framework teams, work with MLIR/LLVM backends, and contribute to features across GPU hardware and software stacks in a fast-paced, production-focused environment.