About the Role
NVIDIA is looking for software engineering interns to join the math libraries team building cuEST, a CUDA-X math library that accelerates quantum chemistry on NVIDIA GPUs. Launched in March 2026, cuEST provides accelerated APIs for essential functions in electronic structure computations, starting with Gaussian-basis quantum chemistry. These capabilities power applications ranging from drug discovery to semiconductor process development. We are constantly working to improve the feature set, performance, and technology stack of cuEST. If the idea of helping expand the frontiers of electronic structure theory excites you, come join our team and help us build GPU-accelerated software used by quantum chemists around the world.
During your internship, you will work with senior engineers on the cuEST team who will mentor you and guide you in developing highly optimized algorithms and numerical methods for quantum chemistry. Projects may involve implementing new quantum chemistry primitives, defining APIs, analyzing and improving GPU performance, finding solutions for difficult numerical corner cases, and other software engineering work that goes into shipping a production CUDA-X library.
What you'll be doing
•Prototype and develop GPU-accelerated algorithms for Gaussian-basis quantum chemistry primitives in cuEST.
•Analyze the performance of CUDA implementations, identify opportunities for improvement, and implement them.
•Contribute across the engineering lifecycle, from design and testing to code that ships in the library.
•Collaborate with team members and other stakeholders to understand how quantum chemistry applications use cuEST and what they need from it.
What we need to see
•Pursuing a BS, MS, or PhD degree in Computational Chemistry, Computational Physics, Computer Science, Applied Math, or an equivalent program area.
•Excellent C/C++ or Python programming, debugging, performance analysis, and test design skills.
•Strong algorithms and numerical methods fundamentals.
•Ability to work independently and lead your own development effort.
•Good teamwork, communication, and documentation habits.
Ways to stand out from the crowd
•GPU programming experience (CUDA or OpenCL) is desired but not required.
•Coursework or research experience in quantum chemistry or electronic structure theory is desired but not required.
•Knowledge of floating-point arithmetic and numerical error analysis.
•Knowledge of CPU or GPU hardware architecture.
•Experience using agentic AI tools to improve the speed and quality of software development.
Our internship hourly rates are a standard pay based on the position, your location, year in school, degree, and experience. The expected hourly rate for this position is 20 USD – 94 USD. You will also be eligible for Intern benefits.
Applications for this job will be accepted at least until November 6, 2026. This posting is for an existing vacancy.
NVIDIA uses AI tools in its recruiting processes.
