Software Engineering Intern

NVIDIA is seeking a Software Engineering Intern to improve the continuous integration systems used to develop and verify our deep learning compilers software.

During the internship, you will own a focused project that makes CI failures faster to detect, understand, and resolve.

Working with a mentor and engineers across compiler verification and infrastructure, you will build an AI-assisted workflow for analyzing test results, classifying failures, identifying regressions, and reducing manual triage.

Responsibilities

• Design and implement an AI-assisted workflow for analyzing and triaging failures in deep learning compilers CI
• Use logs, test results, and execution metadata to classify failures, identify regressions, and detect flaky tests
• Integrate the solution with existing CI infrastructure and create diagnostics that help engineers understand and resolve failures
• Evaluate accuracy, reliability, and developer time savings across representative GPU hardware and software configurations
• Collaborate with compiler verification and infrastructure engineers, document the project, and present results to the team

Requirements

• Currently pursuing a BS, MS, or PhD in Computer Science, Computer Engineering, Electrical Engineering, or a related technical field
• Strong Python programming skills and a solid understanding of software engineering, debugging, and version control
• Familiarity with Linux development environments and command-line tools
• Experience through coursework, research, open-source work, or prior internships with CI/CD, test automation, developer tools, infrastructure, or machine learning systems
• Strong analytical, communication, and collaboration skills, with enthusiasm for improving engineering quality and developer productivity

Nice to Have

• Experience with CI platforms such as Jenkins (groovy), GitLab CI, GitHub Actions, Buildkite, or similar systems
• Experience building AI- or LLM-assisted tools for log analysis, test generation, failure triage, or developer support
• Familiarity with Docker, Kubernetes, Slurm, cloud platforms, or observability tools
• Exposure to GPU computing, CUDA, machine learning frameworks, or compiler technologies such as LLVM, MLIR, XLA/HLO, or Triton