Strong engineering skills, especially in CUDA/Triton/Pallas/CuTe DSL kernel development, lower-level PyTorch/JAX/XLA development, CUDA Graphs, or FPGA/ASIC experience
Two or more years of work experience building deep learning systems in any domain such as language, speech, robotics, biology, chemistry, physics, audio, video, or recommendations
Comfortable writing PyTorch or JAX code
Nice to Haves
Experience translating methods between different application areas
Deep learning research experience
Experience with large language models (LLMs)
Finance experience (not required)
What You'll Be Doing
Improving all aspects of large-scale model training including kernel development, training data loading, parallelism, networking, and fault-tolerance
Working closely with researchers to co-design and improve models and help shape the research agenda
Maximizing performance on large, modern, rapidly growing GPU clusters
Contributing to multiple strategies and model types with direct, clear, and significant business impact
Perks and Benefits
Competitive estimated base salary range of $250,000 to $300,000 USD per year (or local equivalent)
Discretionary performance-based bonuses
Comprehensive benefits package
High GPU-to-researcher ratio and cutting-edge computing environment
Diverse, inclusive, and collaborative company culture with a strong focus on openness, transparency, and togetherness
Equal opportunity employer with global offices and a welcoming community of self-starters