An AI-powered social hub for researchers that transforms dense academic papers into snackable slide-stories, conversational podcasts, and crisp reels. Built for IIIT-H but universally adaptable—making research discoverable, social, and actually fun.
- Multi-Slide Stories: Papers transformed into visual, swipeable stories with Gemini-powered summaries
- Podcast Generation: Convert research into natural multi-person audio discussions with TTS
- Video Reels: Short-form, brainrot-friendly video summaries for the fast-paced crowd
- Smart Summaries: AI-generated abstracts, key findings, and methodology breakdowns
- Reddit-Style Communities: Upvote/downvote discussions, nested replies, and community threads
- Lab Spaces: Dedicated forums for CVIT, LTRC, SERC, RRC, and other research centers
- Stories Feed: Instagram-style research highlights you can consume in seconds
- Universal Search: Find papers, users, posts, reels, podcasts, and discussions in one place
- Personalized Feed: AI-driven recommendations based on your interactions and interests
- Co-Author Requests: Find teammates with specific skills across campus
- User Profiles: Showcase research interests, contributions, and lab affiliations
- Community Voting: Upvote valuable insights, downvote noise—Reddit-style democracy
- Like/Dislike Tracking: Your preferences shape your feed
- Category Interests: Dynamic scoring based on what you engage with
- Interaction History: Track your liked posts, saved papers, and contributions
- Lab Discovery: Explore ongoing work at IIIT-H research centers
- Cross-Lab Synergy: Break down silos between different research domains
- External Papers: Track papers from researchers beyond campus
- Mobile-First Design: Swipe through research on your commute
- Dark Mode: Easy on the eyes during late-night paper sessions
- Bottom Navigation: Instagram-like tab bar for seamless browsing
- Responsive: Works flawlessly across devices
- Docker & Docker Compose
- Gemini API Key (get free from Google AI Studio)
cp .env.example .envEdit .env and add your GEMINI_API_KEY.
docker compose up --build- Frontend: http://localhost:3000
- Backend API: http://localhost:8000
- API Docs: http://localhost:8000/docs
# From project root
docker compose exec backend python -m backend.seedhackiiit-2026-anotheruselessgroup/
├── backend/ # FastAPI Application
│ ├── main.py # Entry point
│ ├── database.py # SQLite + SQLModel config
│ ├── models.py # Database schemas
│ ├── seed.py # Database seeding
│ ├── arxiv_utils.py # ArXiv API integration
│ ├── papers_service.py # Paper scanning & import
│ ├── ai_content_service.py # Gemini integration
│ ├── content_generation_service.py # Multi-slide post generation
│ ├── reel_generator_service.py # Video reel creation
│ ├── audio_processing.py # Podcast TTS generation
│ ├── arxiv_carousel_service.py # Story generation
│ └── routers/ # API endpoints
│ ├── auth.py # Authentication & JWT
│ ├── arxiv.py # ArXiv scanning
│ ├── feed.py # Posts & personalized feed
│ ├── reels.py # Video reels
│ ├── podcast.py # Podcast episodes
│ ├── community.py # Community discussions
│ ├── labs.py # Research lab pages
│ ├── search.py # Universal search
│ ├── recommendation.py # Personalization engine
│ ├── users.py # User profiles
│ └── admin.py # Admin operations
├── frontend/ # Next.js 16 Application
│ └── src/
│ ├── app/ # App router pages
│ │ ├── (auth)/auth/ # Login/Register
│ │ └── (tabs)/ # Main tabs
│ │ ├── page.tsx # Feed
│ │ ├── explore/ # Search
│ │ ├── community/ # Communities
│ │ ├── reels/ # Video reels
│ │ ├── podcasts/ # Audio episodes
│ │ ├── labs/ # Research labs
│ │ ├── profile/ # User profile
│ │ └── admin/ # Admin panel
│ ├── components/ # React components
│ │ ├── feed/ # Stories, posts
│ │ ├── community/ # Voting, threads
│ │ ├── reels/ # Swipeable reels
│ │ ├── podcasts/ # Audio player
│ │ └── explore/ # Search results
│ └── lib/
│ ├── api.ts # API client
│ └── utils.ts # Helpers
├── seed_databank/ # Seed data
│ ├── categories.json # ArXiv categories
│ ├── labs_data.json # IIIT-H labs
│ └── default_user.json # Test users
├── docker-compose.yml # Dev environment
└── docker-compose.prod.yml # Production setup
Backend
- FastAPI (Python 3.11+) - Modern async web framework
- SQLModel - Type-safe ORM with Pydantic validation
- SQLite - Lightweight database (production-ready with proper config)
- Google Gemini API - AI content generation & summarization
- gTTS - Text-to-speech for podcast generation
- arXiv API - Research paper discovery
Frontend
- Next.js 16 - React framework with App Router
- TypeScript - Type safety throughout
- TailwindCSS - Utility-first styling
- Radix UI - Accessible component primitives
- React Query - Data fetching & caching
DevOps
- Docker & Docker Compose - Containerization
- GitHub Actions - CI/CD (optional)
- Nginx - Production reverse proxy (in prod compose)
| Variable | Description |
|---|---|
GEMINI_API_KEY |
Required for AI features |
SQLITE_DB_PATH |
Path to SQLite DB |
SECRET_KEY |
JWT authentication key |
NEXT_PUBLIC_API_URL |
Frontend API target |
We're an open-source project built for HackIIIT 2026! Thus, the work is not a refined piece, and the team welcomes contributions from the community to enhance and expand Hadron's capabilities.
- Fork the repository
- Create a feature branch:
git checkout -b feature/amazing-feature - Make your changes and test thoroughly
- Commit with clear messages:
git commit -m 'Add amazing feature' - Push to your fork:
git push origin feature/amazing-feature - Open a Pull Request
- Follow existing code style (we use Copilot, so it's pretty consistent)
- Add docstrings to new functions
- Test your changes locally with Docker
- Update relevant documentation
- Keep commits atomic and well-described
- Unit tests for backend services
- Mobile app (React Native?)
- Performance optimizations
- Accessibility improvements
- Integration with IIIT-H CAS authentication
Built with caffeine, Copilot, and pure determination during HackIIIT 2026.
Track: Open Track (Campus Improvement)
MIT License - see LICENSE for details.
- OSDG - IIIT Hyderabad for organizing HackIIIT
- arXiv for open access to research papers
- Copilot for AI-assisted coding
⚛️ Where Ideas Collide ⚛️
Made with ❤️ for researchers who deserve better tools