On-device ML Integration Engineer, Graphics, Games & ML

Apple Apple · Big Tech · Cupertino, CA · Machine Learning and AI

This role focuses on integrating ML models into Apple's on-device inference stack, optimizing performance, and ensuring functionality across various Apple devices. It involves working with ML frameworks, compilers, and hardware targets to enable efficient and private AI experiences.

What you'd actually do

  1. Ensure functional and performant integration of Apple’s ML models across the inference stack.
  2. Integrate Apple’s ML tools into internal and external model repositories to demonstrate and stress-test model ingestion with peak efficiency and performance.
  3. Develop optimizations across the pipeline, including model-level transformations, custom operations, or compiler optimizations to improve inference efficiency.
  4. Spearhead the integration of the cutting-edge ML models with peak performance, using these examples to validate or improve Apple’s inference stack.

Skills

Required

  • Python
  • C++
  • PyTorch
  • MLX
  • JAX
  • ML fundamentals
  • Transformers
  • GPU programming paradigms
  • communication skills

Nice to have

  • C++
  • Swift
  • GPU kernel optimizations
  • MLIR/LLVM
  • Hugging Face

What the JD emphasized

  • excellent user experience
  • flawless functionality
  • maximum performance
  • peak efficiency and performance
  • improve inference efficiency
  • peak performance

Other signals

  • on-device ML
  • inference stack
  • model compression
  • acceleration
  • ML compilers
  • runtimes
  • benchmarking
  • debugging toolchains
  • Apple Intelligence ecosystem