Senior ML Infrastructure Engineer - Training Algorithms, SIML

Are you passionate about Generative AI? Are you interested in working on groundbreaking generative modeling technologies to enrich billions of people? We are the Intelligence System Experience (ISE) team within Apple's software organization. The team operates at the intersection of multimodal machine learning and system experiences. Our multidisciplinary ML teams focus on a broad spectrum of areas, including Visual Generative Foundation Models, Multimodal Understanding, Visual Understanding of People, Text, Handwriting, and Scenes, Personalization, Knowledge Extraction, Conversation Analysis, Behavioral Modeling for Proactive Suggestions, and Privacy-Preserving Learning. These innovations form the foundation of the seamless, intelligent experiences our users enjoy every day.\\n\\nWe are seeking engineers experienced in building infrastructure for training, adapting and deploying large-scale generative models. In this role, you will be working with closely with a cross functional team of algorithm design and infrastructure engineers to benchmark, prototype and steer algorithmic choices to best fit our training & deployment infrastructure.

In this role you will be technically hands on, with deep subject matter expertise in ML infrastructure. \n\nResponsibilities Include: \n- Training optimizations & profiling targeting vision/language pre-training\n\n- Researching training recipes for effective scheduling of multimodal training workloads\n\n- Experimentation & tooling for post-training ablations including reward modeling, distillation and prompt optimization\n\n- Coordinating with post-training algorithm owners for analyzing quality / performance tradeoffs of downstream capabilities\n\n- Ablations involving optimization aware fine-tuning

Bachelors, Masters, or PhD in Electrical Engineering/Computer Science or a related field (mathematics, physics or computer engineering), with a focus on machine learning; or comparable professional experience\n\nExperienced in training / adapting LLM and Diffusion models\n\nAdvanced Fluency in PyTorch \n\nExcellent programming skills and experience contributing software to large projects\n\nExperience with distributed training of large models

Strong ML Fundamentals\nExperience working with large cross-functional and diverse teams.