GenAI for Supply Chain at Apple using RAG

I led the end-to-end design and implementation of a Generative AI solution at Apple, developed using a Retrieval-Augmented Generation (RAG) architecture. This system ingests supplier audit findings and intelligently generates tailored Corrective Action Plans (CAPs). My role included designing the technical architecture, training and fine-tuning LLMs, integrating Apple’s audit standards into the system, and deploying the solution within Apple’s SupplierCare portal. I also built feedback loops and compliance review integrations to ensure high accuracy and learning over time.

This solution represents one of the first enterprise-scale applications of Generative AI specifically tailored for supply chain compliance workflows. Prior systems required heavy human intervention and manual documentation. By embedding a RAG-powered architecture that combines audit-specific retrieval with large language model reasoning, this project introduces a novel way to automate judgment-based compliance tasks — a space typically untouched by AI at scale. It is also designed to evolve into an intelligent audit expert, a rare implementation of self-improving GenAI in regulatory environments.