Vector database
VectorPrism — multi-channel 1024d tensor retrieval (dense + relational + disentangled + hyperbolic + identity + causal)
GDPR/CCPA Right-to-be-Forgotten Engine for Vector Databases (ChromaDB, Qdrant, Pgvector) with Global PII Extraction, Dry Run & SHA-256 Audit Trails.
Build a Production RAG System with Python, LangChain, and ChromaDB Build ShopBot from scratch — a production RAG assistant that answers from retrieved evidence, not training memory. Ten chapters, one complete system.
Open-source graph database in Go — Cypher, graph algorithms, vector search (HNSW), ontology inference, multi-tenant. A lightweight Neo4j alternative in a single binary.
基于 RAG 的 AI 智能笔记助手「云尚」：FastAPI + LangGraph + ChromaDB + Redis 构建。ReAct / Plan-and-Execute 双 Agent 模式、深度思考、图片/视频多模态、MCP 联网搜索与网页抓取、DeepL 翻译 / Wolfram 计算 / Edge TTS 语音、PPT 生成、记忆压缩、知识库重排序、笔记回收站与邮件导出、Token 用量追踪。
Adaptive routing strategies for distributed vector databases. Reduces communication costs by 42% while maintaining retrieval quality. Uses TurboVec (ICLR 2026) and DBpedia 14.
Read-only staleness, orphan, duplicate and deleted-but-retrievable checks for pgvector, Qdrant and Chroma indexes.
Memory infrastructure for agentic coding
An automated correlation engine that detects unexpected connections between NASA articles, lunar sensor data, and Apollo astronaut transcripts. Built with Qdrant, SQLite, Sentence Transformers, and Python.
Pure V client for MongrelDB - embedded and server database with SQL, vector search, full-text search, and AI-native retrieval
Coroutine-first Kotlin client for the Qdrant vector database — suspend APIs, type-safe DSLs, small footprint.
LanceDB VS Code Extension for viewing tables and databases
On-device Retrieval-Augmented Generation using Qdrant Edge, Qwen3 embeddings, and Gemma4.
Modular framework with vector database (LanceDB), filesystem archive (Mycelium), MCP server ecosystem, and Blender simulation
Local RAG semantic search for Obsidian vaults — Ollama + LanceDB + MCP. Search your markdown vault by meaning, not keyword.
Persistent, local, cross-IDE memory for AI agents — markdown source of truth, LanceDB-powered semantic search, zero cloud dependency