AI Compiler Engineer: Kernel Gen & MLIR Innovation

NVIDIA; Cambridge; On-site; GBP 110,000 - 160,000; Full time

Job summary

NVIDIA invites applications for an AI Compiler Engineer to push AI performance on next‑generation GPUs. You will work on kernel generation and graph optimizations, tackling compiler challenges for inference and training, and partner with software, hardware, and research teams to co-design future silicon.

A BS/MS in CS/CE (PhD preferred) with 3+ years of compiler experience and hands-on MLIR is required. Proficiency in C/C++ and Python, strong debugging, and collaboration skills are essential.

Qualifications

• BS/MS in CS/CE (or related). PhD strongly preferred.
• 3+ years of compiler optimization, synthesis, and placement experience.
• Hands-on MLIR experience.
• Excellent C/C++ and Python programming and software design, including debugging, performance analysis, and test design.
• Strong communication and collaboration in a fast-paced, product-focused environment.

Responsibilities

• Drive technical innovation focusing on kernel generation and graph optimizations for next-generation NVIDIA GPUs.
• Advance AI workloads compilation for inference and training and transition breakthroughs to products.
• Collaborate on hardware/software co-design with software, hardware, and research teams to architect future silicon.
• Scale AI to datacenter deployments and optimize workloads.

Skills

C/C++

Python

Debugging

Performance analysis

Test design

Communication

Teamwork

Education

BS or MS in Computer Science/Computer Engineering or related field; PhD strongly preferred

Tools

MLIR

Job description

NVIDIA invites applications for an AI Compiler Engineer to push AI performance on next‑generation GPUs. You will work on kernel generation and graph optimizations, tackling compiler challenges for inference and training, and partner with software, hardware, and research teams to co-design future silicon.

A BS/MS in CS/CE (PhD preferred) with 3+ years of compiler experience and hands-on MLIR is required. Proficiency in C/C++ and Python, strong debugging, and collaboration skills are essential.