Senior Machine Learning Compiler Engineer

The AWS Neuron Compiler team is seeking skilled compiler engineers to join our efforts in developing a state-of-the-art deep learning compiler stack for optimizing application models across diverse domains. Your role will involve working with custom-built Machine Learning accelerators, including Inferentia/Trainium, to power solutions like Generative AI. Key job responsibilities: Develop and maintain tooling for the Neuron Compiler's accuracy and reliability Lead the efforts building fuzzers and specification synthesis tooling for our LLVM-based compiler Work in a team with a science focus to push the boundaries of what is known in delivering solutions to customers Strong software development skills using C++/Python are crucial. A science background in compiler development is strongly preferred, and experience in Machine Learning and AI accelerators is a plus. The role is located in Seattle, Cupertino, Austin, or Toronto. BASIC QUALIFICATIONS: 6+ years of experience in developing compiler features and optimizations Proficiency in C++ and Python programming applied to compiler or verification projects Familiarity with LLVM, including knowledge of abstract interpretation and polyhedral domains Demonstrated scientific approach to software engineering problems PREFERRED QUALIFICATIONS: Masters degree or PhD in computer science or equivalent Experience with deep learning frameworks like TensorFlow or PyTorch Understanding of large language model (LLM) training processes Knowledge of CUDA programming for GPU acceleration

Created: 2024-11-05