Articles for people who
ship the stack
Rewrites on React, Node.js, TypeScript and AI. A translated body is shown when it exists. If it does not, you see the English text and a note that says so.
Tagged: SQLite
Context Engineering for a Quant Research Agent: Proof Levels and Data Stores
Context engineering for a WorldQuant research agent: rules, observations and beliefs at separate proof levels, spread across a ledger, event store and graph.
12 min readROE Versus TechRAG: GraphRAG on Docker and Qdrant or on SQLite and LanceDB
Two takes on one GraphRAG thesis: Qdrant, MongoDB and a 120B model against LanceDB, SQLite and qwen2.5 7B, with retrieval logic compared and lessons drawn.
7 min readA Medical Agent That Routes Between SQLite Databases and Web Search
Build a LangChain agent that decides whether a medical question needs a SQL query on three datasets or a live web search, plus two breakages that changed the design.
11 min readBuild an MCP Server in Python and Connect It to Claude Desktop and SQLite
Build an MCP server in Python that wraps a SQLite database, register it with Claude Desktop on Windows, and let the model write and run its own SQL queries.
4 min readClinic Booking Chatbot With LangGraph, SQLite and Streamlit, Step by Step
Build an appointment-booking assistant for a clinic: SQLite tables, service functions, a nine-node LangGraph workflow with gpt-4o-mini, a Streamlit chat UI and a notebook mode.
17 min readAutomating a Daily AI Content Pipeline With a Local Qwen3 Model
How a local Ollama pipeline finds, scores and drafts developer content each day, with a quick human approval step cutting three hours of work to about ten minutes.
4 min readThread Memory for a LangGraph Support Agent With SQLite Checkpoints
How a LangGraph support agent keeps multi-turn context using SqliteSaver and thread ids, and why short-term thread memory differs from long-term memory.
6 min readCodeGraph: Give Your AI Coding Agent a Knowledge Graph of Any Folder
How CodeGraph indexes a codebase with Tree-sitter and SQLite, serves it over MCP and cuts agent tool calls by 70% and token costs by nearly half, all locally.
4 min readLocal RAG Without Vectors: SQLite FTS5 and BM25 for Code Search
Why dense vector indexes strain developer machines, and how SQLite FTS5 with BM25 gives faster, exact-match code retrieval at a fraction of the memory.
3 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.