We build frontier foundation models that power intelligent experiences at Apple. Our team works across the full training lifecycle: including pre-training foundation models, and developing mid-training approaches that bridge general capability and task-specific performance. What makes our work distinct is that we're engineering models specifically for Apple silicon and optimized for experiences that are private, personal, and deeply integrated into the OS. We're solving frontier problems in reward modeling to resist reward hacking, handling sparse and delayed rewards in agentic settings, and aligning models reliably across the spectrum from open-ended creative tasks to precise, action-taking workflows. If you're drawn to hard problems where the research and the product are inseparable, this is the team.

Architect Scalable Ingestion Systems: Design and implement high-throughput distributed systems to ingest petabytes of text and multimodal data from diverse sources, including web crawls and third-party partnerships.

Repository Optimization: Manage the lifecycle of large-scale datasets across data storage and high-performance file systems. Optimize data formats for efficient random access and sequential scanning during model training.

Data Governance & Privacy: Engineer robust data governance and privacy solutions for the training data, in collaboration with compliance and legal teams, to ensure adherence to stringent regulatory standards.

High-Performance Processing Pipelines: Build and maintain distributed data processing workflows using advanced frameworks on cloud infrastructure (e.g., GCP, AWS).

Algorithmic Data Curation: Implement sophisticated data filtering and selection logic to remove low-quality content. Develop semantic deduplication at scale to prevent model memorization and improve training efficiency.

Decontamination Removal: Design automated systems to detect and remove benchmark leakage, ensuring that evaluation datasets remain strictly isolated from training corpora.

Infrastructure for Scaling Laws: Collaborate with researchers to enable data ablations and scaling experiments. Build tools to support systematic data mixture optimization and empirically data studies.