Compiler Verification Engineer, Compute Performance – GPU

Job Title

Compiler Verification Engineer, Compute Performance – GPU

Role Summary

Join the Compiler Verification Team to validate compiler correctness and performance for NVIDIA GPU compute stacks. The role focuses on diagnosing performance regressions and functional defects, automating compiler tests, and maintaining performance baselines to ensure sustained compiler quality.

Experience Level

Mid-level. The posting requests 3+ years of relevant experience in software development or testing.

Responsibilities

Primary responsibilities include investigating compiler performance issues, creating automated tests, and working with development teams to resolve regressions.

• Analyze performance degradations and functional defects in compilers; identify root causes and recommend corrective actions.
• Automate compiler testing: author and integrate tests, run test suites, generate reports, and automate build/test environments.
• Operate test infrastructure: use test suites to detect, report, track, and drive performance regressions to resolution.
• Develop and review test plans and test cases; port and integrate third-party tests into NVIDIA test frameworks.
• Maintain historical performance data and baselines; generate statistics and monitor performance trends.
• Participate in process and quality-improvement initiatives and iterative planning for test development.
• Collaborate with compiler and software engineers to reproduce issues, triage complex regressions, and implement fixes.

Requirements

Must-have skills and experience required to perform the role; nice-to-have items are listed separately.

• Must-have: 3+ years' experience in a software development or test organization.
• Must-have: Strong experience programming and/or testing in C/C++ and familiarity with CUDA; experience with scripting languages (Python, Perl, Shell).
• Must-have: Deep understanding of SDLC, HPC, and software testing methodologies; proven problem-solving and analytical skills.
• Must-have: Experience writing test plans, developing and automating tests, executing tests, and producing test reports in production environments.
• Must-have: Ability to triage complex performance regressions, work under tight schedules, and communicate clearly across teams.
• Nice-to-have: Prior compiler development, compiler verification/test, or performance analysis experience.
• Nice-to-have: Experience with NVIDIA CUDA Toolkit and debugging on Linux.
• Nice-to-have: Familiarity with revision control and project tools (Git, Perforce, JIRA, Confluence, Make).
• Nice-to-have: Experience using statistical analysis tools to identify out-of-bound behavior.

Education Requirements

Bachelor's or Master's degree preferred, or equivalent practical experience. No specific field of study is required or specified in the posting.

About the Company

Company: NVIDIA

Headquarters: Santa Clara, California, USA

NVIDIA is a global leader in accelerated computing, renowned for its innovative solutions in AI and digital twins that transform diverse industries. The company specializes in networking technologies, providing end-to-end InfiniBand and Ethernet solutions for servers and storage that optimize performance and scalability. NVIDIA serves sectors such as high-performance computing, enterprise data centers, and cloud computing, constantly reinventing its products and services to stay ahead in the market.