Applied Scientist Ii, Annapurna ML

Amazon Amazon · Big Tech · Cupertino, CA · Research Science

Applied Scientist II role focused on enhancing ML accelerator software (Trainium/Inferentia) to accelerate customer adoption. Responsibilities include developing ML/RL for code generation/optimization, creating ML compiler techniques, building validation tools, and designing high-performance kernels. The role involves working with customers, engineering teams, and research communities to advance ML systems, with a focus on inference performance and training cost optimization.

What you'd actually do

  1. work directly with external and internal customers to identify key adoption barriers and optimization opportunities
  2. collaborate closely with our engineering teams to implement innovative solutions
  3. engage with academic and research communities to advance state-of-the-art ML systems
  4. Developing and applying ML/RL approaches for kernel/code generation and optimization
  5. Creating advanced compiler techniques for ML workloads
  6. Building tools for accuracy and reliability validation
  7. Designing high-performance kernels optimized for our ML accelerator architectures

Skills

Required

  • PhD, or Master's degree and 4+ years of CS, CE, ML or related field experience
  • Experience programming in Java, C++, Python or related language
  • Experience in any of the following areas: algorithms and data structures, parsing, numerical optimization, data mining, parallel and distributed computing, high-performance computing

Nice to have

  • Experience in state-of-the-art deep learning models architecture design and deep learning training and optimization and model pruning
  • Experience building machine learning models or developing algorithms for business application

What the JD emphasized

  • Experience in patents or publications at top-tier peer-reviewed conferences or journals

Other signals

  • accelerating customer adoption of Trainium and Inferentia accelerators
  • implement innovative solutions
  • advance state-of-the-art ML systems
  • AI for Systems: Developing and applying ML/RL approaches for kernel/code generation and optimization
  • Machine Learning Compiler: Creating advanced compiler techniques for ML workloads
  • System Robustness: Building tools for accuracy and reliability validation
  • Efficient Kernel Development: Designing high-performance kernels optimized for our ML accelerator architectures
  • AWS Neuron is the software of Trainium and Inferentia, the AWS Machine Learning chips
  • Inferentia delivers best-in-class ML inference performance at the lowest cost in the cloud
  • Trainium is designed to deliver the best-in-class ML training performance at the lowest training cost in the cloud
  • Neuron is a Software that include ML compiler and native integration into popular ML frameworks
  • Our products are being used at scale with external customers like Anthropic and Databricks as well as internal customers like Alexa, Amazon Bedrocks, Amazon Robotics, Amazon Ads, Amazon Rekognition and many more.