Description

Design and research high-performance communication technologies and libraries for next-generation GPU and HPC platforms.

This role is on-site.

Responsibilities

• Research new communication technologies and expand the GPUDirect technology portfolio.
• Co-design hardware and software solutions with GPU, networking, and software architects for seamless integration.
• Drive adoption of new communication technologies across application verticals using quantitative data and proof-of-concepts.
• Collaborate with diverse internal and external teams, including DL researchers and customers, to integrate latest DL research.

Required Skills

• PhD in Computer Science, Computer Engineering, or related field, or strong equivalent experience.
• 15+ years of relevant experience in academia or the industry.
• Expertise in HPC, parallel programming models (MPI, SHMEM), and computer/system architecture.
• Deep understanding of GPU architecture, CUDA, and at least one communication runtime (MPI, NCCL, NVSHMEM, OpenSHMEM, UCX, UCC).
• Strong background in high-performance networking, including Infiniband, Ethernet, network topologies, and performance analysis.
• Programming fluency with C or C++ for systems software development.
• Experience with ML/DL fundamentals, parallel algorithms, and performance optimization for large clusters.

Preferred Skills

• Industry recognition as a leader in HPC/DL communications with patents, publications, or conference keynotes.
• Influential role in industry standards (e.g., MPI, OpenSHMEM) or open-source software (e.g., PyTorch, UCX, Open MPI).

Key Skills

Education

Any Graduate