#pgvector

Multi-agent memory substrate for PostgreSQL — provenance-gated, vector-hybrid recall

Local-first MCP server for shared memory, reservations, handoffs, and review evidence across AI coding agents.

Trustworthy AI Knowledge Workbench — Verifiable RAG with citation tracking, document-grounded review, and self-contained desktop mode. Windows / macOS / Linux / Android / iOS.

Agent + Multimodal RAG 全栈学习项目：LangGraph PAE/ReAct、Multi-Query、BM25 + pgvector 混合检索、RAG-Fusion/RRF、MinerU/OCR/ASR、evidence 引用与 HITL；React + Spring Boot + FastAPI。

🚀 Opendray v2 2026 – Universal AI Agent Gateway (Slack, Telegram, Discord, DingTalk)

HermesClaw Zero‑Config Sidecar is a lightweight local vector‑memory engine for Hermes and OpenClaw. It runs as a zero‑config sidecar, providing fast embeddings, persistent memory, and instant retrieval for agents and MCP tools. Built for speed, simplicity, and full local control.

A full-stack equity research platform pairing retrieval-augmented filing analysis with a deterministic quantitative engine — cited document Q&A, DCF valuation, ratio diagnostics, peer benchmarking, sentiment, and portfolio analytics. Built with FastAPI, React, and PostgreSQL/pgvector.

Self-healing RAG on n8n with LLM-as-judge eval, drift detection, investigator agent, and canary A/B deploys.

AI-помощник по документации: Go + Next.js + pgvector + Ollama

Self-hosted memory for Claude — markdown knowledge base with wikilink graph, hybrid semantic search (pgvector + Ollama), and a first-party MCP server. One docker compose up.

Human-in-the-loop equity research agent: SEC-filing RAG with cited evidence, deterministic risk guardrails, and live execution through Robinhood's Trading MCP

Open-source enterprise AI agent platform for centrally governing agents, tools, memory, RAG, approvals, and durable workflows with FastAPI, LangGraph, LangChain, and LangSmith.

Document Q&A chatbot built with FastAPI, Supabase pgvector, and Groq — answers strictly from the uploaded document, with Docker support.

LangGraph-based Supervisor-Worker multi-agent framework with a modern web GUI.

Masar AI — agentic RAG over Dubai RTA open data. A 14-agent LangGraph orchestration with a corrective loop, hybrid retrieval (pgvector + Postgres FTS), Text-to-SQL, and deterministic fare arithmetic. Bilingual AR/EN, fully cited, runs on free-tier/local inference. Independent academic project, not affiliated with RTA.

My latest flagship project — a multi-tenant RAG customer-support SaaS. Upload your docs, embed one line of code, and let AI answer customer questions with citations. Built with Next.js 16, React 19, FastAPI, PostgreSQL + pgvector, Voyage embeddings, Clerk auth, and Anthropic Claude.

Agentic pipeline that classifies router logs, filters noise, detects anomalies, and generates engineer-readable diagnoses using LangGraph, Groq, and pgvector.

Enterprise-grade agentic RAG chatbot — Python 3.14 / FastAPI / LangGraph orchestration, PostgreSQL + pgvector hybrid (dense + BM25) retrieval with cross-encoder reranking, Keycloak OIDC auth, MinIO object storage, arq ingestion workers, DeepEval scoring, and a pixel-perfect React 19 + Vite + TypeScript UI.

Persistent memory for coding agents over MCP — typed knowledge, semantic search, session lifecycle. PostgreSQL + pgvector, FastMCP.

Spring Boot microservices backend for a professional networking platform, featuring an agentic AI "Content Studio" (Spring AI + RAG + LLM) that plans, drafts, self-critiques, moderates, and publishes posts. Built with Java 21, Kafka, PostgreSQL/pgvector, Neo4j graph database, and Eureka.

Chat with your documents — production-grade RAG on Spring Boot 4 + Spring AI 2.0 + pgvector, exposed as both a REST API and an MCP server for Claude Desktop, with eval-gated CI and LLM cost observability.

Multi-tenant AI customer support SaaS — .NET 10, React, PostgreSQL + pgvector

Export a personal Slack channel, enrich + classify it with Claude, and build a Cognee knowledge graph — a showcase for the community pgGraph adapter.

AI career-matching agent for Lebanese CS graduates - semantic job retrieval, deterministic skill-gap analysis and a measurable Precision@5 benchmark.

Deploy a Discord LLM chatbot with persistent sessions, memory and semantic search in one click.

Generic, provider-agnostic RAG backend with verified citations. FastAPI + SSE; switchable LLMs (Ollama/Anthropic/OpenAI) and vector stores (Chroma/pgvector); evals, tracing, tests, Docker.

Dynamic episodic memory for LLMs. Self-hosted Rust microservice. Vector index + temporal graph + tiered retrieval + REST/MCP.