Remote AI/HPC Middleware Engineer

Cornelis Networks; United States; On-site; USD 120,000 - 180,000; Full time

Job summary

Cornelis Networks is seeking a full-time remote Software Engineer to work on AI/HPC middleware, developing and optimizing communication systems for AI and HPC workloads. You will focus on enabling and tuning features, and integrate with transport layers while contributing to upstream open-source projects.

You will implement and optimize features for HPC middleware (MPI/SHMEM) and AI stacks (CCL such as NCCL/RCCL), collaborating on libfabric/OFI, UCX, and contributing upstream to related projects.

Qualifications

• 2+ years of systems programming experience on Linux in C/C++.
• Exposure to HPC middleware (MPI/SHMEM) and/or AI collectives (NCCL/RCCL).
• Familiarity with performance concepts and diagnostic tools.
• Foundational understanding of networking and/or RDMA concepts.

Responsibilities

• Implement and optimize features for HPC middleware (MPI/SHMEM) and AI middleware (CCL stacks such as NCCL/RCCL).
• Assist in integrating middleware capabilities with transport and provider layers (e.g., libfabric / OFI, UCX).
• Prepare and submit upstream contributions across MPI/SHMEM projects and CCL ecosystems.

Skills

C/C++ programming

Linux development

Performance analysis

Education

Bachelor's degree in CS/Engineering or related field

Tools

libfabric/OFI

UCX

MPI

SHMEM

NCCL

RCCL

Job description

Cornelis Networks is seeking a full-time remote Software Engineer to work on AI/HPC middleware, developing and optimizing communication systems for AI and HPC workloads. You will focus on enabling and tuning features, and integrate with transport layers while contributing to upstream open-source projects.

You will implement and optimize features for HPC middleware (MPI/SHMEM) and AI stacks (CCL such as NCCL/RCCL), collaborating on libfabric/OFI, UCX, and contributing upstream to related projects.