At AWS AI, the business problem is to make deep learning pervasive for everyday developers and democratize access to cutting-edge infrastructure, specifically through the development of the AWS Neuron SDK for the AWS Inferentia and Trainium machine learning accelerators.
Requirements
- Deep background in Linux and device drivers
- Experience with hardware-based accelerator platforms (such as encryption, TCP offload, storage accelerators, and router/switch platforms)
- Experience implementing high-performance software drivers
- Experience programming with at least one software programming language
- Experience in computer architecture
- Experience writing low-level drivers
- Experience with full software development life cycle, including coding standards, code reviews, source control management, build processes, testing, and operations
Responsibilities
- Define the key interfaces for the AWS Neuron SDK
- Develop the runtime stack and driver that manages the accelerator
- Implement high-performance software drivers to ensure high throughput from software to hardware
- Work with the team to deliver high-performance and cost-effective inference and training
- Develop and manage the compiler, runtime, frameworks, and tooling for the AWS Neuron SDK
- Ensure high standards for performance, quality, and cost of the products
- Collaborate with thought-leaders in multiple technology areas to improve products
Other
- 3+ years of non-internship professional software development experience
- 2+ years of non-internship design or architecture (design patterns, reliability, and scaling) of new and existing systems experience
- Bachelor's degree in computer science or equivalent
- Ability to work in a fast-paced, intellectually challenging environment
- High standards for oneself and others
