Explores how enterprise AI context windows need explicit memory, reset, and retrieval boundaries to prevent drift and errors.
Explains how to design governed RAG systems using data products, separating retrieval and governance for accurate, policy-compliant AI responses.
Explains how rerankers improve search and AI results by reordering retrieved documents for better precision and relevance.
Argues that building a good search engine is more critical for effective RAG than just using a vector database, as poor retrieval misleads AI.
An experiment comparing retrieval performance of chunked vs. non-chunked documents using long-context embedding models like BGE-M3.
Explains how to implement document retrieval with the Azure OpenAI Assistants API using a custom RAG approach, as the retrieval tool is not yet natively supported.
A technical guide exploring the OpenAI Assistants API, covering its core concepts and demonstrating how to create an assistant with the Code Interpreter tool.
A developer shares experiments building LLM-powered tools for research, reflection, and planning, including URL summarizers, SQL agents, and advisory boards.
