o4

ESG knowledge graph + RAG agent. Turn long-form sustainability disclosures into cited answers, extracted entities, and a reviewable graph of evidence.

Persistent knowledge graph memory for AI agents — Neo4j + vector search + NL→Cypher. Agent-agnostic: Hermes, Claude Code, LangChain.

Local-first GraphRAG: Milvus hybrid (dense+BM25) + Neo4j local-search + cross-encoder rerank + contextual retrieval. Local-dev runs end-to-end on Ollama with no paid APIs; CI integration tier uses Ollama Cloud + OpenAI embeddings.

Skein is an agentic AI platform that answers questions by weaving together structured and unstructured data, grounding every response in a knowledge graph, hybrid vector search, and a governed semantic layer. It escalates to humans when unsure, and runs free locally or scales to the cloud.

(Cooking...) Production-grade Multi-Agent Personal Operating System built with LangGraph, FastAPI, PostgreSQL, Redis, Qdrant, Neo4j, MCP, OAuth, and Long-Term Memory.

agent wayfinding. temporal knowledge store in one sqlite file. no neo4j, no cloud, no lock-in.

Luminary — voice-first deep-research agent. Talk to it, it runs a multi-phase pipeline (query analysis → study planning → iterative source-grounded research → claim validation → synthesis) and reads the result back. Google ADK + Gemini + OpenAI o4-mini + ElevenLabs on Cloud Run.

SceneThread is a video intelligence agent that watches raw video, extracts key scenes, entities, objects, speech, and events, then connects them into a searchable Neo4j context graph. It lets users ask natural-language questions and trace how people, objects, locations, and actions relate across the video.

CPF Customer Agent - Powered by Neo4j & GraphRAG

Claude-powered RAG with citations, GraphRAG over a Neo4j knowledge graph, and side-by-side Claude vs GPT-4.

An enterprise-grade, hybrid Retrieval-Augmented Generation (RAG) platform powered by LangGraph, Google Vertex AI (gemini-2.5-flash, text-embedding-004), PostgreSQL (pgvector), and Neo4j 5.20.

GraphRAG assistant for the EU AI Act (Regulation (EU) 2024/1689), built on a Neo4j knowledge graph. Ask compliance questions in natural language - the agent routes between graph traversal, vector search, or both, answers with article citations, and visualizes the exact path it traversed through the graph.

基于 Agent 的智能学习与知识管理平台，通过 RAG 混合检索 + 知识图谱 + 多智能体协作，实现"知识输入 → 知识理解 → 学习规划 → 学习执行 → 能力评估 → 个性化优化"的全流程智能化管理。支持文档上传与感知型解析（扫描件多模态理解）、知识实体抽取与图谱构建、个性化学习计划、AI 导师流式问答、四维练习评测与错题复盘、能力画像与学习预测。后端 FastAPI + Neo4j + FAISS + Celery，前端 React + Ant Design.

Production-grade Enterprise AI Operating System — multi-agent orchestration via LangGraph, hybrid RAG with Qdrant + BM25, real-time SSE streaming chat with LLaMA 3.3 70B (Groq), Neo4j knowledge graph, async document ingestion, RBAC auth, and an observability dashboard. Built with FastAPI & Next.js.

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.

LegalMind 是一个由 AI Agent 驱动的企业级法律合同审查平台。它通过多 Agent 并行协作与长链推理（Chain-of-Thought），自动识别高风险条款、交叉引用冲突、跨境合规漏洞，并生成可落地的修改建议与谈判策略。让合同审查时间从数小时缩短至数分钟，同时显著提升风险检出率。

agenticflow synchromodal transport