Senior Distributed AI Training Engineer

Genesis AI; Paris; On-site; EUR 90,000 - 150,000; Full time

Job summary

Genesis AI in Paris is seeking a senior ML infrastructure engineer to cut wall-clock time in foundation model training by profiling bottlenecks and optimizing the full stack from data pipelines to GPU kernels.

You will design distributed PyTorch training across multi-node GPU clusters, implement low-level CUDA/Triton code, and ensure scalable, robust, and highly utilized systems. Strong Python production skills and a system-minded approach are essential.

Qualifications

• 8+ years of experience in distributed systems, ML infra, or HPC.
• Proficient in Python, production-grade coding practices.
• Expertise in CUDA/cuDNN/Triton and kernel optimization.
• Experience with PyTorch distributed training on multi-node GPUs.
• Strong system-level approach to optimize CPU-GPU interactions.

Responsibilities

• Profile and eliminate bottlenecks to reduce wall-clock time in training stacks.
• Design, build, and optimize distributed PyTorch training across multi-node GPUs.
• Implement low-level code and integrate into high-level training frameworks.
• Optimize hardware utilization across CPU/GPU, memory, data throughput, and networking.
• Develop monitoring and debugging tools for large-scale runs.

Skills

Distributed systems

ML infrastructure

High-performance computing

Python

PyTorch

Kernel optimization

CPU–GPU optimization

Tools

CUDA

cuDNN

Triton

PyTorch

Job description

Genesis AI in Paris is seeking a senior ML infrastructure engineer to cut wall-clock time in foundation model training by profiling bottlenecks and optimizing the full stack from data pipelines to GPU kernels.

You will design distributed PyTorch training across multi-node GPU clusters, implement low-level CUDA/Triton code, and ensure scalable, robust, and highly utilized systems. Strong Python production skills and a system-minded approach are essential.