Training / AI Infrastructure

Genesis AI; Paris; On-site; EUR 90,000 - 140,000; Full time

Job summary

Genesis AI in Paris seeks a senior ML infrastructure engineer to optimize distributed training stacks across data pipelines and GPU kernels. You will design, build, and optimize multi-node PyTorch systems for scalable, robust performance.

Responsibilities include implementing low-level CUDA/cuDNN/Triton code, balancing CPU-GPU workloads, and developing monitoring tools to diagnose performance regressions, ensuring high hardware utilization in large-scale runs.

Qualifications

• 8+ years in distributed systems, ML infrastructure, or HPC.
• Production-grade Python expertise.
• Deep CUDA/cuDNN/Triton knowledge and kernel optimization.
• Experience with PyTorch multi-node training and data, context, pipeline, and model parallelism.

Responsibilities

• Profile bottlenecks across data pipelines, GPU kernels, and training stacks.
• Design and optimize distributed training systems for multi-node GPUs.
• Implement low-level CUDA/CuDNN/Triton code and integrate with PyTorch.
• Tune CPU-GPU balance, memory management, data throughput, and networking.
• Develop monitoring and debugging tools for large-scale runs.

Skills

Distributed systems

ML infrastructure

High-performance computing

Python production-grade

PyTorch training

System-level tuning

Tools

CUDA/cuDNN/Triton

Job description

What You’ll Do

• Drive down wall-clock time to convergence by profiling and eliminating bottlenecks across the foundation model training stack stack, from data pipelines to GPU kernels
• Design, build, and optimize distributed training systems (PyTorch) for multi-node GPU clusters, ensuring scalability, robustness, and high utilization
• Implement efficient low-level code (CUDA, cuDNN, Triton, custom kernels) and integrate it seamlessly into high-level training frameworks
• Optimize workloads for hardware efficiency: CPU-GPU compute balance, memory management, data throughput, and networking
• Develop monitoring and debugging tools for large-scale runs, enabling rapid diagnosis of performance regressions and failures

What You’ll Bring

• Deep experience in distributed systems, ML infrastructure, or high-performance computing (8+ years)
• Production-grade expertise in Python
• Low-level performance mastery: CUDA/cuDNN/Triton, CPU-GPU interactions, data movement, and kernel optimization
• Scaling at the frontier: experience with PyTorch and training jobs using data, context, pipeline, and model parallelism
• System-level mindset with a track record of tuning hardware-software interactions for maximum utilization