Category Rag

AI & LLM, Rag

Spring AI RAG Debugging: When Retrieved Documents Do Not Reach the Model

Spring AI RAG flow showing retrieved document text entering the model prompt

Debug Spring AI RAG when retrieved documents do not reach the model. Trace source labels, filter replacement, empty results and prompt context with Java 17.

Peng
2026-09-30

AI & LLM, Rag

Text Embeddings for RAG: From Token Lookup to Retrieval Quality

How do tokens become useful retrieval vectors? This article explains embedding lookup, contextual encoding, and retrieval training, then uses a small experiment to explore ranking quality, encoding pitfalls, and model evaluation.

Peng
2026-09-03

AI & LLM, Rag

Making RAG Reliable: Document Quality, Retrieval, and Evidence in a Customer Support Assistant

Reliable RAG requires more than vector search. This article explores how document quality, access controls, policy versions, hybrid retrieval, and citation checks help a customer support assistant produce better-supported answers.

Peng
2026-08-30