Lead Research Engineer
Lightning AI is seeking a Lead Research Engineer to enhance the performance of training and inference workloads on compute accelerators and clusters. This role focuses on the development of the Lightning Thunder compiler and the broader PyTorch Lightning ecosystem. The position is based in one of our hubs in New York City, San Francisco, or London, with a requirement of at least two in-office days per week.
The Lead Research Engineer will develop performance-oriented model optimizations at various levels, including graph-level optimizations such as operator fusion and memory planning, kernel-level enhancements using CUDA or Triton, and system-level improvements for distributed training across GPUs and TPUs. Responsibilities also include advancing the Thunder compiler by building optimization passes and integrating hooks to accelerate workloads, designing profiling tools to analyze model execution, and collaborating with hardware vendors to ensure efficiency across diverse backends.
Candidates should have strong expertise with deep learning frameworks like PyTorch, hands-on experience with model optimization techniques including quantization and mixed precision, and knowledge of distributed systems and parallelism strategies. Familiarity with software engineering practices such as API design and CI/CD for performance-sensitive systems is also required. A Bachelor's degree in Computer Science, Engineering, or a related field is necessary.
Preferred qualifications include experience with GPU programming models like CUDA or Triton, a deep understanding of deep learning compiler internals, a proven track record in open-source contributions, and an advanced degree in machine learning, compilers, or systems.
Lightning AI offers competitive compensation packages, including comprehensive health plans with significant premium contributions, flexible unlimited paid time off, generous paid family leave, monthly meal reimbursements, home office build-out stipends, annual learning and development stipends, and access to wellness programs. The company emphasizes a balanced work-life environment and provides opportunities for professional growth and development.