GenAI Staff Machine Learning Engineer, Performance Optimization

You will:

• Explore and analyze performance bottlenecks in ML training and inference
• Design, implement and benchmark libraries and methods to overcome aforementioned bottlenecks
• Build tools for performance profiling, analysis, and estimation for ML training and inference
• Balance the tradeoff between performance and usability for our customers
• Facilitate our community through documentation, talks, tutorials, and collaborations
• Collaborate with external researchers and leading AI companies on various efficiency methods

We look for:

• Hands on experience the internals of deep learning frameworks (e.g. PyTorch, TensorFlow) and deep learning models
• Experience with high-performance linear algebra libraries such as cuDNN, CUTLASS, Eigen, MKL, etc.
• General experience with the training and deployment of ML models
• Experience with compiler technologies relevant to machine learning
• Experience with distributed systems development or distributed ML workloads
• Hands on experience with writing CUDA code and knowledge of GPU internals (Preferred)
• Publications in top tier ML or System Conferences such as MLSys, ICML, ICLR, KDD, NeurIPS (Preferred)