Description
We are looking for an enthusiastic software engineer to join our AI networking acceleration team, to work on a groundbreaking open-source library, using hardware offloads, GPU Kernels and RDMA network cards.
Our product is a performance-oriented low-level infrastructure, crafted to change the way inference works.
You will be developing a highly optimized inference framework, running on the world’s largest supercomputers and data centers.
The work environment is dynamic and challenging as our employees work on innovative, next-generation products at the forefront of technology in terms of performance, scalability, and features.
Responsibilities:
- Developing a highly optimized inference framework
- Running on the world’s largest supercomputers and data centers
Requirements:
- B.Sc. or equivalent experience in Computer Science or Software Engineering
- 1+ years’ experience in modern C++ / C / Rust development
- 1+ years’ experience in a Linux environment and familiarity with development tools
- Deep knowledge of the TCP/IP network stack
- Understanding of computer architecture and operating systems concepts
Nice to Have:
- Background in Linux internals and low-level software optimizations (benchmarking, bottleneck research, performance tuning)
- Experience in programming CUDA kernels
- Familiarity with ML frameworks and LLMs
- Background in parallel programming / high-performance computing / RDMA technology
NVIDIA is committed to fostering a diverse work environment and proud to be an equal opportunity employer.