Genesis AI
Training / AI Infrastructure
Bay Area · On-site · Full-time · Posted Apr 30, 2026
- Experience
- 8+ years in distributed systems, ML infrastructure, or high-performance computing.
About the job
Build and optimize distributed foundation-model training infrastructure to improve convergence speed, scalability, and hardware utilization.
What you'll do
- Profile the training stack and eliminate bottlenecks across data pipelines and GPU kernels.
- Design and optimize distributed PyTorch training systems for multi-node GPU clusters.
- Implement low-level kernels and build tools to monitor and debug large-scale training runs.
What they ask for
- Deep experience in distributed systems, ML infrastructure, or high-performance computing.
- Production-grade expertise in Python.
- Expertise in CUDA/cuDNN/Triton, CPU–GPU interactions, data movement, and kernel optimization.
- Experience with PyTorch and training jobs using data, context, pipeline, and model parallelism.
Tools
PyTorch, Python, CUDA, cuDNN, Triton
AI focus
Optimize infrastructure for foundation-model training.
Model-written summary · the original posting is authoritative