ML Engineer, Foundation Models

About the Role

Join the Mosaic team at Databricks to train and serve open foundation models. You will work on large-scale distributed training, develop efficient pre-training and fine-tuning recipes, and build the ML infrastructure that powers enterprise AI workloads on the Databricks platform.

Requirements

• MS/PhD in Computer Science or Machine Learning
• 4+ years of experience training large neural networks
• Deep expertise in PyTorch and distributed training (FSDP, DeepSpeed)
• Experience with GPU cluster management and optimization
• Strong systems engineering background

Skills