About the roleWe're looking for seasoned ML Infrastructure engineers with experience designing, building and maintaining training and serving infrastructure for ML research.
Responsibilities:
- Provide infrastructure support to our ML research and product
- Build tooling to diagnose cluster issues and hardware failures
- Monitor deployments, manage experiments, and generally support our research
- Maximize GPU allocation and utilization for both serving and training
Requirements:- 4+ years of experience supporting the infrastructure within an ML environment
- Experience in developing tools used to diagnose ML infrastructure problems and failures
- Experience with cloud platforms (e.g., Compute Engine, Kubernetes, Cloud Storage)
- Experience working with GPUs
Nice to have- Experience with large GPU clusters and high-performance computing/networking
- Experience with supporting large language model training
- Experience with ML frameworks like Pytorch/TensorFlow/JAX
- Experience with GPU kernel development
