SQLite

DuckDB + Parquet for Ad‑hoc Analytics from SQLite

A developer describes a lightweight analytics pattern: populate a single SQLite DB from the YouTube Data API, then nightly export the analytical subset to hive-partitioned Parquet using the DuckDB CLI orchestrated by a small PHP script. The Parquet snapshots are partitioned by snapshot_date and region, compressed with zstd, and pulled to a local machine over FTP (lftp mirror). Local ad‑hoc analysis runs against the Parquet files with DuckDB (CLI or Python API), delivering sub-second queries across months of history; results (JSON) are pushed back to production hosts for simple cached rendering. The post documents implementation details, example SQL/PHP/Python snippets, performance characteristics, and operational gotchas (type casts, timezone normalization, idempotent writes, and partitioning best practices).

Monlite: SQLite-based local stack for AI agents

Monlite is a TypeScript library that consolidates a local AI agent stack into a single SQLite .db file, providing a document store, vector search, full-text search, key-value cache, job queue, and cron scheduler. It uses SQLite's FTS5 and the sqlite-vec extension (vec0 virtual table) for KNN vector queries, supports Node (node:sqlite or better-sqlite3) and a Python port for cross-language interoperability, and implements exactly-once job claiming using SQLite's BEGIN IMMEDIATE write-intent lock. The project targets single-machine, local-first workflows (not high-concurrency distributed systems) and is available on GitHub; the core is at v2.6.1 with a frozen API.

Build a Natural Language Text-to-SQL Database Assistant

A developer tutorial (published on 2026-06-27) demonstrates how to build a Text-to-SQL natural language database assistant using a lightweight Python stack. The author explains leveraging a prefine‑tuned T5 model (t5-base-finetuned-wikiSQL) via the Hugging Face Inference API to translate English questions into executable SQL, then executing those queries against an in-memory SQLite database using pandas and rendering results in a Streamlit UI. The post includes sample code for calling the Hugging Face endpoint, describes the app’s data layer and execution flow, and links to an open-source GitHub repository with the full project. The tutorial frames Text-to-SQL as a practical approach to data democratization, enabling non-technical users to query databases without writing SQL.