An intelligent system that generates contextual, well-defined hackathon problem statements using the LLaMA model. The system crawls relevant sources, analyzes themes, and generates implementable technical challenges.

• Theme-based Generation: Support for multiple domains including AI/ML, Climate Tech, Healthcare, etc.
• Intelligent Analysis: Uses LLaMA model to analyze and generate relevant problems
• Multi-source Data: Aggregates data from arXiv, GitHub, tech blogs, and other sources
• Problem Validation: Ensures problems are practical and implementable
• Difficulty Estimation: Auto-categorizes problems into Easy/Medium/Hard
• Persistent Storage: JSON-based problem database with deduplication

1. Clone the repository:

git clone https://github.com/Prashithshetty/hackbuddy.git
cd api

2. Install dependencies:

pip install -r requirements.txt

3. Download LLaMA model:

• Download DeepSeek-R1-8b.gguf from hugging face or ollama or lmstudio
• Place it in folder ai/
• Update path in line 574 in lama.py

api/
├── lama.py          # Main application logic
├── database.py      # Database handler for problem storage
├── scraper.py       # Multi-source problem scraper
├── requirements.txt # Project dependencies
└── problems_db.json # Problem database

Run the main script:

python lama.py

Follow the interactive prompts to:

1. Select a theme
2. Wait for data collection and analysis
3. Review generated problem statement
4. Problems are automatically saved to database

• AI & ML
• Climate Tech
• Healthcare
• Innovation
• FinTech
• Logistics
• Sustainability

Each problem includes:

• Title
• Technical Challenge Description
• Context and Current Trends
• Impact Assessment
• Technical/Business Constraints
• Success Criteria
• Difficulty Rating

• Python 3.8+
• LLaMA model support
• Internet connection for data scraping
• Minimum 8GB RAM recommended

• llama-cpp-python >= 0.2.0
• beautifulsoup4 >= 4.12.0
• aiohttp >= 3.9.0
• cachetools >= 5.3.0
• requests >= 2.31.0
• feedparser >= 6.0.10
• html2text >= 2020.1.16
• urllib3 >= 2.1.0

• Logs are stored in hackathon_generator.log
• Failed scraping attempts are gracefully handled
• Fallback prompts when data collection fails

1. Fork the repository
2. Create a feature branch
3. Commit your changes
4. Push to the branch
5. Create a Pull Request

Prashith R Shetty

• LLaMA model community
• arXiv API
• GitHub API