AI Infrastructure Engineer - Distributed Training & Kernels

Genesis in Paris, France, is seeking an experienced ML infrastructure engineer to accelerate large-scale model training. You’ll optimize distributed PyTorch pipelines and design high-performance runtimes across multi-node GPU clusters, delivering scalable, robust training workflows.

You’ll implement low-level kernels with CUDA/Triton, tune CPU–GPU interactions, and develop monitoring tools, ensuring efficient hardware utilization.