Posted at: 14 April
Java Compiler Engineering Intern - 2026
Company
NVIDIA Corporation is a Santa Clara-based technology company specializing in designing GPUs and AI solutions for gaming, professional visualization, and cloud services, operating in both B2B and B2C markets globally.
Remote Hiring Policy:
NVIDIA supports flexible remote work arrangements and hires from various regions globally, including the Americas, Europe, Asia, and the Middle East, with roles that may require collaboration across time zones.
Job Type
Full-time
Allowed Applicant Locations
China
Job Description
We are now seeking a Compiler Engineer Intern for an exciting and fun role in our Java Compiler Team. Our team is responsible for improving the JVM that realise the potential of NVIDIA's CPUs designed for the world's largest Data Centers: https://www.nvidia.com/en-gb/data-center/grace-cpu/, which covers the areas over runtime, compiler optimization, code generation, CPU pipeline analysis and etc.What you'll be doing:Work on feature enablement and optimization of open source Java virtual machine projects.Partner with both internal and communities to develop, valid and upstream patches.Work with global compiler, hardware and application teams to oversee improvements and problem resolutions.Drive and push architecture neutral and NV friendly solutions in the communities.Be flexible, with a variety of software development skills and extend the breadth and depth of knowledge.What we need to see:Pursuing B.S. or higher degree in Computer Science/EngineeringExcellent hands-on C++ programming skillsStrong background in software engineering principles with a focus on crafting robust and maintainable solutions to challenging problemsGood communication and documentation skills and self-motivatedWays to stand out from the crowd:Masters or PhD preferred, with compiler or language virtual machine experiencesBackground with Java and understand the key language featuresFamiliar with Linux and shell programmingKnowledge on computer architecture, ISA, assembly, and Arm64 is preferredExposure to various ML techniques