Tech Stack

Description

You will optimize training and inference workloads on compute accelerators and clusters using the Lightning Thunder compiler and PyTorch Lightning ecosystem, working at the intersection of deep learning research, compiler development, and large-scale system optimization.

Requirements

• Strong expertise with deep learning frameworks such as PyTorch
• Hands-on experience with model optimization techniques including graph-level optimizations, quantization, pruning, mixed precision, or memory-efficient training
• Knowledge of distributed systems and parallelism strategies (data/model/pipeline parallelism, checkpointing, elastic scaling)
• Familiarity with software engineering practices: designing APIs, building robust tooling, testing, CI/CD for performance-sensitive systems
• Excellent collaboration and communication skills

Responsibilities

• Develop performance-oriented model optimizations at graph, kernel, and system levels
• Advance the Thunder compiler with optimization passes, graph transformations, and integration hooks
• Ensure optimizations are accessible through clean APIs and seamless integration with PyTorch Lightning
• Design profiling and debugging tools to analyze model execution and guide optimization strategies
• Collaborate with hardware vendors and contribute to open-source projects