Nachiket Mehendale·Oct 13Basic Retrievers in RAG with QdrantChecking Retrieved Chunks Against a Similarity Threshold
Nachiket Mehendale·Oct 8Grouped Search in Qdrant Vector DatabaseA step-by-step tutorial on grouped search feature, using a chunked PDF, Qdrant Vector DB and LangChain.
Nachiket Mehendale·Sep 29The Role of Magnitude in Euclidean Distance and Other Similarity Search MetricsCosine similarity is widely used in RAG pipelines to compare vectors. Euclidean distance (L2) can be a better fit when the magnitude of a…
PrajnaAI·Aug 3Similarity Is Not Relevance: Why Enterprise RAG Needs More Than Vector SearchVector search is remarkably good at finding content that looks like a question.
Vasu Sharma·Jul 10Anisotropic Quantisation in Vector SearchI recently used Scalable Nearest Neighbour Search (ScaNN) for vector search. It works blazingly fast on millions of high dimensional…
Mohamed Orabi·Jun 30raglib: The Clean-Architecture Python Library for Building Production RAG Systems and AI AgentsStop duct-taping LLM providers, vector stores, and retrieval pipelines together. raglib gives you one unified interface, graph-based…A response icon1A response icon1
Krisna Renaldi·Jun 22How to “catch” news using embedding and similarityDi tulisan sebelumnya, embedding dan hitung cosine similarity digunakan untuk mendapat topik berita. Dan sekarang kita coba untuk…
Syed M·Jun 19You Probably Don’t Need a Vector Database — Grep Might Be Good EnoughPlain text search vs BM25 vs embeddings: where each one wins (and where it wastes your time)
Narges Abdolkarimi·Jun 10We All Drew the Same LogoSome brands borrowed a specific mark. The rest of us just dissolved into the category. Either way: cover the name, and there’s nothing left…A response icon11A response icon11
Rahulpaul·Jun 2Postcards from a Data Analysis JourneyThree countries, three dialects of meaning, one curious tourist.
