Tagged articles
FTS5
10 articles · Page 1 of 1
WeChat stores 1.3 billion users' chat records locally on devices using SQLite via its WCDB framework, avoiding EB-scale server costs while enabling instant search via FTS5 full-text indexing, with SQLCipher encryption and automatic sharding at 240MB per database file.
ACIDFTS5SQLCipher
0 likes · 16 min read

This article explains why WeChat stores chat records locally using SQLite instead of cloud databases, detailing the cost savings, privacy benefits, and performance advantages of embedded databases, along with WeChat's custom WCDB framework, FTS5 full-text search, encryption, and the trade-offs of local-only storage.
ACIDFTS5SQLCipher
0 likes · 16 min read

This article argues that SQLite, often dismissed as a toy database, can replace specialized systems like Elasticsearch, MongoDB, Kafka, ClickHouse, Redis, and even microservices due to its stability, zero-configuration deployment, built-in full-text search, JSON support, vector extensions, and local-first architecture, reducing operational complexity.
CachingFTS5JSON
0 likes · 25 min read

This article dissects Hermes Agent’s four‑layer cross‑session retrieval system, covering persistent storage, dual‑table FTS5 indexing for CJK and English, a three‑path search strategy, intelligent truncation for LLM prompts, structured summarisation, and a holographic retrieval layer that blends FTS5, Jaccard similarity and HRR vector algebra.
Agent ArchitectureCross-Session RetrievalFTS5
0 likes · 25 min read


ArchitectApr 13, 2026 · Artificial Intelligence
The article analyzes Hermes Agent's three‑layer memory system—fact memory stored in tiny Markdown files, session history indexed with SQLite + FTS5, and procedural memory via skill management—then compares each layer to OpenClaw's architecture and explains how to integrate self‑summarizing skills into OpenClaw.
Agent ArchitectureExternal Memory ProviderFTS5
0 likes · 27 min read


CodeTrendApr 11, 2026 · Artificial Intelligence
Hermes Agent introduces a closed‑loop learning architecture that adds result evaluation, pattern extraction, and persistent user modeling to the traditional receive‑plan‑execute‑return cycle, offering searchable FTS5‑based memory, autonomous skill creation, multi‑platform messaging, provider‑agnostic model switching, and built‑in research tools for AI developers.
FTS5Hermes AgentLLM summarization
0 likes · 8 min read

The Cloud Music team integrated a lightweight on‑device full‑text engine using SQLite FTS5 with a simple tokenizer, replaced JavaScript matching with SQLite’s bm25(), parallelized queries, and cut search latency by 75%, boosting CTR 13% and average playback by 17 seconds while preserving user privacy.
FTS5SQLitefull-text search
0 likes · 15 min read

SQLite 3.42.0, the widely used public-domain SQL database engine, introduces features such as FTS5 secure-delete, JSON5 support, new sqlite3_config options, enhanced query planner optimizations, CLI improvements, additional SQL functions, extended PRAGMA integrity_check, subsecond time modifiers, and various bug fixes and performance enhancements.
Database EngineFTS5JSON5
0 likes · 5 min read

This article details the 2021 overhaul of iOS WeChat's full‑text search, covering engine selection, segment‑merge optimization, a new VerbatimTokenizer, multi‑level separator support, table schema choices, asynchronous index updates, and extensive performance gains across chat, contacts, and favorites.
FTS5SQLitefull-text search
0 likes · 27 min read

This article explains how WeChat Android tackles the challenge of Chinese homograph full‑text search by using SQLite FTS5, comparing word‑based and character‑based indexing schemes, introducing a custom tokenizer, and detailing preprocessing and performance results.
FTS5SQLiteWeChat
0 likes · 11 min read
