Senior CPU Compiler Engineer – HPC

Nvidia seeks a Senior CPU Compiler Engineer - HPC to join its team in 2 Locations on a full-time basis. This opportunity centers on shaping high-performance computing software that enables superior CPU efficiency and scalability for demanding workloads.

Position Summary

The Senior CPU Compiler Engineer - HPC position at Nvidia in 2 Locations involves crafting compiler technologies that maximize computational throughput in high-performance environments. In this full-time role, you will help refine code generation pipelines and performance tools that support large-scale scientific and engineering applications, working closely with peers to deliver robust, production-ready solutions for complex CPU architectures.

Key Responsibilities

• Develop and refine backend optimizer passes that improve instruction scheduling, register allocation, and loop transformations for high-performance computing codes.
• Profile and tune compiler-generated assembly to reduce latency and increase floating-point throughput on modern multi-core processors.
• Collaborate with architecture and runtime teams to expose new CPU features such as advanced vector extensions and out-of-order execution hints through the compiler stack.
• Diagnose and correct subtle bugs in intermediate representations and target-specific code emitters under real-world HPC workloads.
• Prototype and integrate research-level techniques for automatic parallelization and data-locality improvements that benefit large-scale simulations.
• Maintain high standards of code quality, documentation, and continuous-integration testing for compiler components used by internal and external customers.

What We're Looking For

• Substantial hands-on experience designing or enhancing production CPU compilers, with deep familiarity of popular frameworks.
• Proficiency in C++ and low-level systems programming, including performance analysis with hardware counters and simulators.
• Solid understanding of HPC concepts such as distributed memory models, vectorization, and cache-conscious algorithm design.
• Prior work optimizing generators for scalar, SIMD, or multi-threaded execution on contemporary server-class processors.
• Strong problem-solving skills and the ability to operate autonomously while contributing to cross-functional technical discussions in a full-time setting at Nvidia locations spanning 2 Locations.

About Nvidia

Nvidia is hiring for this role and offers a supportive team environment.

Applying for This Role

To apply, complete your application directly on this page, or you'll be redirected to the employer's application platform to finish submitting there.