GenAI Research Engineer

Apple•Cupertino, CA

About The Position

Join Apple’s Intelligent System Experiences team, which develops core modeling technology behind Genmoji, Image Playground, and Image Wand. The team works across web-scale datasets, distributed foundation-model training, application-specific adapters, and on-device optimization to deliver generative AI experiences to Apple users. The Research Engineer will improve the speed, size, and cost efficiency of production generative models running on devices and servers. The role spans the efficiency stack, including developing and deploying inference optimizations for diffusion models, designing hardware-aware architectures within compute and memory constraints, and owning work from prototyping and evaluation through deployment in the Apple ecosystem.

Requirements

  • M.S. or Ph.D. in Computer Science or a related field, such as Electrical Engineering, Robotics, Statistics, or Applied Mathematics, or equivalent experience.
  • Deep, demonstrated expertise in at least one of the following: quantization, including low-bit and mixed-precision methods or quantization-aware training; efficient architecture design; structured pruning; or neural architecture search for GenAI or vision applications.
  • Strong hands-on experience training and fine-tuning diffusion or other large generative models using PyTorch, JAX, or an equivalent framework.
  • Ability to reason about efficiency beyond FLOPs, including memory, latency, and numerical precision, and willingness to learn hardware-specific constraints.
  • Excellent cross-functional collaboration skills, including the ability to explain model trade-offs to hardware engineers, hardware constraints to researchers, and integration needs to software teams, in writing and in person.

Nice To Haves

  • Publications at CVPR, ICCV, NeurIPS, ICML, ICLR, or MLSys on efficient generative models, or significant open-source contributions in this area.

Responsibilities

  • Develop and deploy algorithms to speed up inference of production diffusion models, including post-training quantization, quantization-aware training, kernel fusion, and model distillation.
  • Design hardware-aware architectures that meet compute and memory constraints while minimizing quality loss.
  • Own the process from prototyping and evaluation through deployment into the Apple ecosystem.
© 2026 Teal Labs, Inc
Privacy PolicyTerms of Service