Applied Scientist, AWS Neuron Science Team
Core
Enhancing the software stack for AWS Trainium and Inferentia accelerators to accelerate customer adoption of ML hardware.
Role type
Applied Scientist (ML Systems & Compilers)
Builds
ML compilers, high-performance kernels, and system robustness tools for AWS ML accelerators.
Domain
Cloud computing, Machine Learning hardware, Compiler technology
Deliverable
production ML models | infrastructure
Required skills
algorithms and data structures, numerical optimization, parallel and distributed computing, high-performance computing, Java, C++, Python, Unix/Linux
Preferred skills
deep learning model architecture design, deep learning training and optimization, model pruning, MxNet, Tensor Flow
Technologies
Trainium, Inferentia, MxNet, Tensor Flow
Responsibilities
Identify key adoption barriers and optimization opportunities with customers, collaborate with engineering teams to implement innovative solutions, engage with academic and research communities, develop ML/RL approaches for kernel/code generation, create advanced compiler techniques for ML workloads, design high-performance kernels optimized for ML accelerator architectures
Seniority
Senior, hands-on IC