A personalized research intelligence platform that helps researchers discover, organize, and interact with academic papers & repositories using AI-powered chat, semantic search, and automated trend analysis.
Quick Start • Features • Screenshots • CLI Tool • API Reference • Architecture • Contributing
RRI (Research & Repository Intelligence) is a full-stack, self-hosted platform designed for researchers, engineers, and teams who want to automate the discovery and analysis of academic papers, open-source repositories, and AI/ML community discussions — all in one place.
RRI continuously collects data from 11+ sources, processes it with NLP pipelines, indexes everything into a vector database for semantic search, and provides an AI chat interface (RAG) so you can ask questions about your research corpus in natural language.
| Audience | Use Case |
|---|---|
| 🎓 Researchers | Track new papers in your field, discover related work, get AI-generated summaries |
| 👩💻 ML Engineers | Monitor trending GitHub repos, HuggingFace models, and community discussions |
| 🏢 Research Teams | Centralized knowledge base with document chat, bookmarks, and weekly digests |
| 📊 Tech Leads | Tech radar, trend analysis, and automated intelligence reports |
|
🔎 Semantic Search across 10,000+ papers & repos with vector similarity |
🤖 Chat with your documents — upload PDFs, DOCX, or ingest GitHub repos |
|
📡 Auto-collect from 11+ sources (ArXiv, GitHub, HuggingFace, OpenReview...) |
💻 Built-in CLI — run OSINT tasks directly from the terminal |
|
📊 Trending & Analytics — Tech Radar, trend charts, community insights |
📚 Personal Library — bookmarks, folders, weekly research digests |
RRI automatically collects and aggregates data from 11+ academic and developer sources:
| Source | Type | Data Collected |
|---|---|---|
| 🔬 ArXiv | Papers | Pre-prints with abstracts, categories, authors |
| 📚 Semantic Scholar | Papers | Citations, references, influence scores |
| 🌐 OpenAlex | Papers | Open-access metadata, concepts, institutions |
| 💻 Papers With Code | Papers + Code | Paper-code links, benchmarks, tasks |
| 🐙 GitHub | Repositories | Stars, forks, languages, topics, README |
| 🤗 HuggingFace | Models + Papers | Model cards, downloads, daily papers |
| 📝 OpenReview | Peer Reviews | ICLR/NeurIPS reviews, ratings, decisions |
| 🟠 Hacker News | Discussions | AI/ML posts, scores, comments |
| ✍️ Dev.to | Blog Posts | Technical articles, tags, reactions |
| 🐘 Mastodon | Social Posts | Research community discussions |
| 🔗 Lemmy | Forum Posts | Federated community discussions |
- Vector-based search powered by Qdrant and BGE embeddings
- Search across papers and repositories simultaneously
- Relevance scoring with percentage match display
- Filter results by type (Papers / Repos / All)
- Retrieval-Augmented Generation pipeline with context-aware answers
- Dual LLM support: Ollama (local, private) + OpenAI GPT-4o (cloud)
- Document Chat: Upload PDFs, DOCX, PPTX → ask questions about your documents
- Repo Ingestion: Ingest entire GitHub repositories via gitingest → chat about code
- Full context mode vs. RAG retrieval mode per conversation
- Conversation history with multi-turn support
- Papers Overview: Category distribution (donut chart), yearly publication trends (bar chart)
- Trending Papers & Repos: Filterable by period (day/week/month), category, language
- Tech Radar: Auto-generated technology trend analysis
- HuggingFace Dashboard: Model rankings, download stats, task distribution
- Community Keywords: Trending topics across platforms with keyword analysis
- Bookmarks & Folders: Organize papers and repos into custom folders
- My Library: Personal document collection with folder tree
- Weekly Reports: Auto-generated research digest summaries
- Paper-Code Linking: Automatically match papers to their implementations
- Citation Enrichment: Bulk update citation counts from Semantic Scholar
- JWT-based authentication with user registration/login
- Per-user document libraries, bookmarks, and conversations
- Role-based access to AI chat features
- Dark/Light theme toggle with smooth transitions
- Responsive design with glassmorphism effects and micro-animations
- Interactive charts built with Recharts
- Knowledge graph visualization with react-force-graph-2d
- Global search bar with keyboard shortcut (
/)
RRI includes a built-in CLI for running OSINT tasks from the terminal. See the full CLI documentation.
rri collect arxiv --query "LLM" --category cs.AI --max-results 100
rri search vector "multi-modal RAG" --limit 10
rri analyze paper 2401.12345 --cloud
rri export report --period weekly --format md
rri chat # Interactive RAG chat
| Technology | Purpose |
|---|---|
| FastAPI | Async REST API framework |
| SQLAlchemy 2.0 | Async ORM with PostgreSQL |
| Celery | Distributed task queue |
| Qdrant | Vector similarity search engine |
| Sentence Transformers | BGE text embeddings |
| Ollama | Local LLM inference (Llama 3) |
| OpenAI API | Cloud LLM (GPT-4o) |
| Alembic | Database migrations |
| Pydantic v2 | Data validation & settings |
| Technology | Purpose |
|---|---|
| Next.js 14 | React framework (App Router) |
| TypeScript | Type-safe JavaScript |
| TailwindCSS | Utility-first styling |
| Recharts | Data visualization charts |
| react-force-graph-2d | Knowledge graph visualization |
| Lucide React | Icon library |
| Axios | HTTP client |
| Technology | Purpose |
|---|---|
| Docker Compose | Multi-container orchestration |
| PostgreSQL 16 | Relational database |
| Redis 7 | Caching & Celery message broker |
| Qdrant | Vector embeddings storage |
| Ollama | Self-hosted LLM runtime |
- Docker & Docker Compose (v2.0+)
- Git
- (Optional) GitHub Personal Access Token for higher API rate limits
- (Optional) OpenAI API key for cloud LLM features
git clone https://github.com/nhdandz/ResearchRover.git
cd ResearchRover
cp .env.example .env
# Edit .env with your API keysmake upThis launches 8 containers: app, worker, beat, postgres, redis, qdrant, ollama, frontend.
make migrate # Run database migrations
make pull-model # Download Ollama LLM model
make seed # (Optional) Seed demo data| Service | URL |
|---|---|
| 🌐 Frontend | http://localhost:3000 |
| ⚡ Backend API | http://localhost:8000 |
| 📖 API Docs | http://localhost:8000/docs |
| 🔍 Qdrant | http://localhost:6333/dashboard |
📚 For detailed setup, environment variables, and local development: see Configuration Guide
- 🔔 Real-time alerting with email/Slack notifications
- 📈 Advanced trend analysis with time-series visualization
- 🌍 Multi-language support (Vietnamese paper sources already integrated)
- 📱 Mobile-responsive PWA
- 🔗 BibTeX export and Zotero integration
- 🧩 Plugin system for custom data sources
- 📊 Comparative analysis dashboards
- 🤝 Team collaboration features
| Document | Description |
|---|---|
| CLI Tool | Full CLI reference with all commands and options |
| API Reference | REST API endpoints, authentication, examples |
| Architecture | System design, data pipeline, project structure |
| Configuration | Environment variables, local dev setup, testing |
| Deployment | Docker deployment, Cloudflare Tunnel, VPS guide |
| Contributing | How to contribute, code style, PR process |
This project is licensed under the MIT License — see the LICENSE file for details.
Built with ❤️ for the research community
If you find RRI useful, consider giving it a ⭐ on GitHub!