About the Role
We are looking for talented System Developers and Researchers to join the Snowflake AI Research team and contribute to LLM inference and training system development, optimizations, and agentic systems. Our mission is to build the most efficient and scalable generative AI systems. Recent releases from our team include SwiftKV, an advanced inference optimization, and Arctic LLM, one of the largest open-source MoE foundation models. This is an exciting opportunity to collaborate with a world-class team, including founding members of DeepSpeed, vLLM, and TensorFlow. Together, we will push the boundaries of deep learning systems and drive cutting-edge innovations in AI.
Responsibilities
- Analyze and optimize GPU kernel performance for training and inference of LLMs.
- Develop and implement strategies to enhance the efficiency and scalability of deep learning systems.
- Profile and benchmark deep learning systems using tools and techniques to identify bottlenecks.
- Design and implement optimizations to reduce latency and improve resource utilization for training and inference.
- Stay updated with the latest advancements in GPU kernel optimization, deep learning, and LLM system development.
- Contribute to the development of agentic frameworks and applications for LLM-driven workflows, enhancing automation, reasoning, and decision-making capabilities.
- Open-source and publish innovations, optimizations, and engineering practices in technical blogs, top-tier conferences and journals.
Requirements
- Bachelor’s degree in Computer Science, Electrical Engineering, or a related field. A Master’s degree or PhD is preferred.
- 5 years of experience in GPU kernel optimization, deep learning system optimization, or high-performance computing (HPC).
- Proficiency in deep learning frameworks such as PyTorch, TensorFlow, JAX.
- Strong understanding of GPU architectures and experience with CUDA or similar frameworks.
- Experience with frameworks like CUTLASS, Triton, cuDNN, etc.
- Experience with profiling tools (e.g., nvprof, Nsight) and performance analysis methodologies.
- Solid problem-solving skills and ability to debug complex performance issues.
- Excellent communication skills and ability to work effectively in a cross-functional team environment.