Generative AI Research Engineer

Quick Facts

• Role: Research Engineer focused on GenAI efficiency

Description

You will help make generative models dramatically faster, smaller, and cheaper to run on-device and on server. You will work across the full efficiency stack, including inference optimizations for production diffusion models, hardware-aware architecture design within compute/memory limits, and end-to-end ownership from prototyping and evaluation to deployment.

Responsibilities

• Develop and deploy optimization algorithms to speed up inference of production diffusion models (e.g., PTQ, QAT, kernel fusion, model distillation)
• Design hardware-aware architectures that fit within compute and memory limits with minimal quality loss
• Own the end-to-end process from prototyping and evaluation to deployment into the Apple ecosystem

Requirements

• M.S. or PhD in Computer Science or a related field, or equivalent experience
• Deep, demonstrated expertise in at least one of: quantization (low-bit and mixed-precision, QAT) or efficient architecture design (structured pruning or neural architecture search) for GenAI/Vision
• Hands-on experience training and fine-tuning diffusion or other large generative models using PyTorch, JAX, or equivalent frameworks
• Ability to reason about efficiency beyond FLOPs (memory, latency, numerical precision) and learn hardware-specific constraints
• Excellent cross-functional collaboration skills to explain trade-offs and constraints across research, hardware, and software teams

Benefits

• Opportunity to advance and ship state-of-the-art technology that impacts Apple users globally
• Work across web-scale datasets, distributed training, and on-device optimizations to build production-ready efficiency improvements