Senior AI Middleware Engineer for High-Performance Training

Cornelis Networks; Austin (TX); On-site; USD 120,000 - 150,000; Full time

Benefits offered by this job

Equity, cash, and incentives

Health coverage

401(k) with company match

Open Time Off

Paid holidays

Job summary

Cornelis Networks is seeking experienced engineers in Austin, Texas, to work on cutting-edge networking solutions for AI and HPC. The role focuses on performance-critical feature implementation, distributed training optimization, and profiling workloads. Candidates should possess over 8 years of experience in high-performance systems programming with strong knowledge of GPU communication stacks. The position is remote for U.S. residents and offers a competitive compensation package, including health and retirement benefits.

Qualifications

• 8+ years of experience in high-performance systems programming in C/C++ on Linux.
• Strong experience with GPU communication stacks including CUDA/ROCm and NCCL/RCCL.
• Ability to optimize distributed training performance using profiling and tracing.
• Experience delivering production-quality code.

Responsibilities

• Design and implement performance-critical features for CCL enablement.
• Optimize distributed training performance across multi-node, multi-GPU configurations.
• Improve GPU communication paths including GPU-direct transfers.
• Profile distributed AI workloads and identify bottlenecks.

Skills

High-performance systems programming in C/C++

GPU communication stacks including CUDA/ROCm

Optimizing distributed training performance

Open-source contributions

Tools

AI frameworks such as PyTorch Distributed

CUDA

NCCL

Job description

Cornelis Networks is seeking experienced engineers in Austin, Texas, to work on cutting-edge networking solutions for AI and HPC. The role focuses on performance-critical feature implementation, distributed training optimization, and profiling workloads. Candidates should possess over 8 years of experience in high-performance systems programming with strong knowledge of GPU communication stacks. The position is remote for U.S. residents and offers a competitive compensation package, including health and retirement benefits.