The Engineering Mindset — Experiences, Insights & Ideas

Introduction In my previous post, I introduced Retrieval-Augmented Fine-Tuning (RAFT) — an approach that helps Large Language Models (LLMs) learn how to use external knowledge sources dynamically. In this post, I’ll show you how to build a prototype RAFT stack on AWS — using 100% managed services: ✅ Amazon S3 → Document storage✅ Amazon OpenSearch…

Introduction Large Language Models (LLMs) such as GPT, Claude, and LLaMA have transformed how we interact with information. However, they suffer from two key limitations: Enter Retrieval-Augmented Fine-Tuning (RAFT) — a hybrid approach that combines the best of fine-tuning and Retrieval-Augmented Generation (RAG). RAFT enables LLMs to learn how to intelligently use external knowledge during…