Articles for people who
ship the stack
Original rewrites on React, Node.js, TypeScript and AI — practical notes from the same engineering practice behind our operator software. Article bodies are in English.
Tagged: Embeddings
Index-Time RAG Quality: Chunk Types, Cleanup and Retrieval Labels
Index-time RAG quality: build text, table and image chunks, scrub parser noise, and add LLM-written retrieval labels that roughly halved failed lookups.
12 min readWhy RAG Exists: Parametric Memory, Dense Retrieval and the 2020 Paper
Why RAG exists: the limits of parametric memory, how BM25 and embeddings led to dense retrieval, and what RAG-Sequence and RAG-Token trade off in the paper.
7 min readSemantic Search Service with a Pinecone Vector Database Behind an MCP Server
A reusable semantic search service: chunk PDFs, store embeddings in a Pinecone vector database, and let any agent query them through a FastMCP server.
7 min readBuild a Local RAG Pipeline in Python with No API Costs
A hands-on tutorial building retrieval-augmented generation on CPU with sentence-transformers, numpy and Flan-T5, including chunking, search and an A/B test.
12 min readA Practical RAG Architecture on Firebase and Vertex AI with Firestore Vectors
RAG on Firebase with Vertex AI embeddings and Firestore vector search: batching, metadata filters, tenant isolation, grounded prompts, caching and metrics.
10 min readWhy RAG Still Gives Wrong Answers: Hybrid Search, Reranking and Recall@K
Clean chunks don't guarantee good answers. See where vector retrieval fails and how fusion, a cross-encoder reranker, HyDE and Recall@K measurement fix it.
6 min readLayer by Layer: Designing a Real RAG Retrieval Pipeline for AI Agents
Plan a RAG pipeline with SourceDocument and DocumentChunk contracts, chunking, embeddings and ChromaDB, all behind a RetrievalService that keeps the agent stable.
10 min readHybrid Image Search at Scale: SigLIP, SPLADE, YOLOv8 and Gemini Captions
Design of a million-image search system that pairs dense and sparse embeddings with object detection, entity recognition, safe search and LLM-written descriptions.
6 min readHow RAG Gives an LLM Your Private Data: Chunks, Vectors, Prompts
Understand retrieval-augmented generation from ingestion to query, with a small Node.js example, three common failure modes and a RAG versus fine-tuning table.
8 min read
About these articles
Request a 24h estimate
Need the same stack in a production operator layer? Send the brief — estimate within 24 hours.