Position Details
About this role
This role involves developing and optimizing GPU communication libraries for high-performance computing systems, focusing on multi-node GPU networking and open-source contributions.
Key Responsibilities
- Support AMD's RCCL; Design, implement, and test networking features; Benchmark, profile, and optimize code; Collaborate with technical experts; Deploy libraries on large clusters
Technical Overview
The technical environment includes C, C++, Python, Linux Kernel, GPU networking protocols like RoCE, Libfabric, InfiniBand, and HPC benchmarking tools, primarily working within AMD's ROCm ecosystem.
Ideal Candidate
The ideal candidate is a senior software engineer with extensive experience in GPU communication libraries, proficient in C, C++, and Python, with a strong background in high-performance computing and open-source development. They should have hands-on experience with GPU networking technologies like RoCE, Libfabric, and InfiniBand, and be capable of developing and optimizing multi-node GPU communication systems.
Must-Have Skills
Nice-to-Have Skills
Tools & Platforms
Required Skills
Hard Skills
Soft Skills
Industry & Role
Keywords for Your Resume
Deal Breakers
Lack of experience with GPU communication libraries, No experience with Linux Kernel or open-source contributions, No background in high-performance computing, Unable to work in a hybrid environment
Get matched to jobs like this
Luna finds roles that fit your skills and career goals — no endless scrolling required.
Create a Free Profile