https://www.m365.fm/entity/reranking/Reranking is a retrieval step in which a second model rescores the top candidates returned by a search so the most relevant results are passed on, for example to a language model in a RAG pipeline.enhttps://www.m365.fm/community/u/mirko-p/blog/why-pure-vector-search-fails-enterprise-ai-and-how-to-fix-it/https://www.m365.fm/community/u/mirko-p/blog/why-pure-vector-search-fails-enterprise-ai-and-how-to-fix-it/Wed, 26 Aug 2026 13:09:05 -0000articleWhen you start deploying generative AI across your organization, you quickly realize that standard Retrieval-Augmented Generation (RAG) setups lean heavily on pure …https://www.m365.fm/show/episodes/the-billion-vector-problem-hnsw-vs-diskann-in-azure-ai-search/https://www.m365.fm/show/episodes/the-billion-vector-problem-hnsw-vs-diskann-in-azure-ai-search/Mon, 08 Jun 2026 00:00:00 -0000episodeThe explosion of AI-powered applications has created a new challenge: how do you efficiently search through billions of vector embeddings without exploding infrastructure costs? In this episode, we explore the “billion-vector problem” and compare two leading…https://www.m365.fm/show/episodes/i-engineered-copilot-for-35-million-pages-the-epstein-files-challenge/https://www.m365.fm/show/episodes/i-engineered-copilot-for-35-million-pages-the-epstein-files-challenge/Sun, 07 Jun 2026 00:00:00 -0000episodeIn this episode, Mirko Peters explores one of the most ambitious experiments in enterprise AI: engineering Microsoft Copilot to analyze and reason across more than 35 million pages of highly complex and interconnected information, inspired by the scale and…https://www.m365.fm/show/episodes/vector-search-is-not-a-strategy-the-new-standard-for-copilot-accuracy/https://www.m365.fm/show/episodes/vector-search-is-not-a-strategy-the-new-standard-for-copilot-accuracy/Fri, 01 May 2026 00:00:00 -0000episodeThis episode explains that vector search on its own is not a complete or reliable strategy for building accurate AI systems like Copilot. While many organizations rely on embeddings and vector databases to retrieve information, this approach often leads to…https://www.m365.fm/show/episodes/the-azure-ai-foundry-trap-why-most-fail-fast/https://www.m365.fm/show/episodes/the-azure-ai-foundry-trap-why-most-fail-fast/Mon, 03 Nov 2025 00:00:00 -0000episodeIn this episode we walk through what really happens when Azure AI Foundry doesn’t behave the way you expect, especially when the Agent Service or deployments start acting up. Azure AI Foundry is supposed to feel smooth, almost invisible, tying together OpenAI…