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
RAG Document Assistant with Testable Services and Next.js Chat
Split HTTP from pure Python services, index documents, cite segments, and ship a split-panel UI without forking retrieval.
4200 wordsRead articleChunking is the whole retrieval system: Spring AI 800 vs 128 trap
When the splitter emits 800-token chunks but the embedder only reads 128, retrieval cannot see the truncated tail. Chunk size, window, and overlap are the retrieval system.
3727 wordsRead articleToken Budgets vs. Semantic Boundaries: Getting RAG Chunk Splits Right
Learn how tokenization and chunking differ, where each sits in a RAG pipeline, and how to split documents by meaning while measuring size with the real tokenizer.
2515 wordsRead articleRetrieval by Association: A Memory-Based Mental Model for Vector Search
Learn how embeddings, semantic similarity, approximate nearest neighbor search and metadata filters work together, using human memory as a guide.
2091 wordsRead articleFifteen LLM Concepts Traced Through a Single Support-Bot Request
Follow one customer question through an AI assistant to learn what models, tokens, embeddings, context, RAG, agents and evaluation really do, and where each one stops.
3094 wordsRead articleWhy an LLM Call Is Not an App: Where LangChain Fits in a RAG Pipeline
Learn what LangChain actually does by tracing a document question-answering app from PDF upload to grounded answer, and see when an alternative framework fits better.
2551 wordsRead articleMapping RAG: Pipeline Stages, Core Components and the Variant Landscape
Understand how retrieval-augmented generation works end to end, which components a RAG system needs, and how the many named RAG variants fit into one map.
2771 wordsRead articleChoosing and Tuning Embedding Models for Production RAG Systems
Learn how embedding models turn text into searchable vectors, why domain vocabulary breaks semantic search, and how to select, compress, and fine-tune models for production RAG.
6591 wordsRead articleHow Vector Databases Really Work: From Embeddings to Hybrid Search
Explains how embeddings encode meaning, how similarity search and indexing scale, and when hybrid search and vector databases actually fit enterprise AI systems.
1864 wordsRead article
About these articles
Request a 24h estimate
Need the same stack in a production operator layer? Send the brief — estimate within 24 hours.