About the job
We are looking for a Director, Machine Learning to join our AI/ML Model Compiler and Applications team. If you are excited by the challenge of distributed training of large models on Neural Processing Unit (NPUs), and if you are passionate about improving training efficiency while innovating and generating new ideas, then this role is for you. You will be part of a world class team focused on addressing the challenge of training generative AI.
Responsibilities
Lead a growing team focused on research, design, and implement novel methods for efficient Generative AI.
Experience in developing ML compilers for low power inference architecture
Experience in all 3 stages of ML compiler- front end, middle end and backend
Experience in defining optimized tiling algorithms, fusion algorithms at different memory hierarchy
Influence the direction of AMD AI platform.
Collaborate across teams with various groups and stakeholders.
Propose and apply innovative model quantization, sparsity, and acceleration algorithms on various AMD platform, e.g., GPU/CPU/AIE.
Collaborate with software and hardware team to E2E co-optimize model performance.
Work with open-source framework and community (e.g., Pytorch, Huggingface) to integrate AMD optimized models and libraries.
Qualifications
Minimum
No minimum qualifications listed.
Preferred
Experience with ML frameworks such as Onnx, PyTorch, or TensorFlow.
Experience with distributed training and distributed training frameworks, such as DeepSpeed.
Experience with LLMs or computer vision, especially large models, is a plus.
Excellent C++ and Python programming skills, including debugging, profiling, and performance analysis.
Experience with ML pipelines.
Strong communication and problem-solving skills.
Strong leadership and management skills.