Post-Training Research Engineer

📋 Description

• Build in-house tooling to support post-training models
• Work across the stack: Kubernetes, storage, networking
• Leverage PyTorch distributed tensor computation and GPU kernels
• Train a wide spectrum of model architectures at scale
• Collaborate with researchers to define specs and requirements
• Address systems-level ML infra and tooling challenges

🎯 Requirements

• Deep understanding of modern ML techniques for training transformers
• Advanced experience with PyTorch, TensorFlow, or Jax
• Knowledge of transformer training parallelism: data, tensor, pipeline
• Ability to profile and optimize distributed GPU programs
• Familiarity with HPC and distributed platforms: Slurm, Ray, Kubernetes, Dask
• Familiarity with cluster networking: Infiniband, RoCE, GPUDirect

🎁 Benefits

• Competitive compensation, including meaningful equity
• 100% medical/dental/vision coverage for employees and dependents
• Generous PTO including Winter Break
• Paid parental leave
• Company-facilitated 401(k)
• Exposure to a variety of ML startups and learning opportunities