ML Compiler Engineer, Annapurna Labs

Amazon Amazon · Big Tech · Cupertino, CA · Software Development

Seeking a skilled compiler engineer to develop and optimize a deep learning compiler stack for AWS ML accelerators (Inferentia/Trainium), supporting frameworks like PyTorch, TensorFlow, and JAX for domains including LLMs and Vision. Responsibilities include compiler feature design, optimization, and collaboration with hardware, runtime, and framework teams.

What you'd actually do

  1. designing, developing, and optimizing features for our compiler
  2. tackling crucial challenges alongside a talented engineering team, contributing to leading-edge design and research in compiler technology and deep-learning systems software
  3. collaborate closely with cross-functional team members from the Runtime, Frameworks, and Hardware teams to ensure system-wide performance optimization
  4. designing and developing various aspects of our system. This includes but is not limited to instruction scheduling, memory allocation, data transfer optimization, graph partitioning, parallel programing, code generation, Instruction Set Architectures, new hardware bring-up, and hardware-software co-design
  5. Solve challenging technical problems, often ones not solved before, at every layer of the stack

Skills

Required

  • optimization algorithms
  • graph-theory
  • hardware bring-up
  • C++
  • compiler engineering

Nice to have

  • PyTorch
  • TensorFlow
  • JAX
  • LLM
  • Vision

What the JD emphasized

  • state-of-the-art deep learning compiler stack
  • custom-built Machine Learning accelerators
  • forefront of AWS innovation for advanced ML capabilities
  • leading-edge design and research in compiler technology
  • new hardware bring-up

Other signals

  • ML accelerators
  • deep learning compiler stack
  • optimize application models
  • Generative AI