Submission Guidelines

Submission Deadline

Final submissions are due: 5:00 PM (CT), Sept 12

All projects must be submitted by this deadline to be eligible for judging and prizes.

Submission Requirements

1. Project Repository

Your project must include a public GitHub repository with:

• Clear README with project description and setup instructions
• Well-documented code with comments
• Requirements file (e.g., requirements.txt, environment.yml)
• License file (recommended: MIT, Apache 2.0, or similar open-source license)

2. Project Documentation

Include the following in your repository:

README.md Must Contain:

• Project title and brief description
• Problem statement - What challenge are you addressing?
• Solution approach - How does your project use LLMs?
• Installation and setup instructions
• Usage examples with sample inputs/outputs
• Team members and their contributions
• Acknowledgments and data sources used

Additional Documentation:

• Technical details - Architecture, algorithms, model choices
• Results and evaluation - Performance metrics, validation
• Future work - Potential improvements and extensions
• References - Related work and citations

3. Demonstration Materials

Provide clear evidence of your project’s functionality:

Required:

• Working code that can be executed by judges
• Example outputs - Screenshots, generated files, or results
• Demo video (2-3 minutes) showing your project in action

Optional but Recommended:

• Live demo - Hosted application or interactive notebook
• Presentation slides - For final presentations
• Test data - Sample datasets for validation

Submission Process

Step 1: Prepare Your Repository

1. Ensure your GitHub repository is public and accessible
2. Complete all required documentation
3. Test that others can reproduce your results
4. Add appropriate tags/releases for the hackathon submission

Step 2: Submit Your Project

Submission form will be available at: Submission Portal

Provide the following information:

• Team name and member details
• GitHub repository URL
• Project category (see categories below)
• Brief project summary (250 words max)
• Demo video URL (YouTube, Vimeo, or similar)

Step 3: Final Presentations

• Selected teams will present their projects on 12, 2025
• Presentations will be 5 minutes + 2 minutes Q&A
• Both virtual and in-person presentations accepted

Project Categories

Choose the category that best fits your project:

🔬 Materials Discovery & Design

• Novel material generation and optimization
• Property prediction and materials screening
• Synthesis pathway planning

⚗️ Chemical Research & Analysis

• Reaction prediction and mechanism analysis
• Molecular design and drug discovery
• Chemical safety and risk assessment

📊 Data Processing & Visualization

• Automated literature review and synthesis
• Data extraction and knowledge graphs
• Interactive visualization tools

🤖 AI Assistants & Automation

• Research assistants for scientists
• Automated experimental design
• Scientific writing and documentation tools

🔧 Tools & Platforms

• Domain-specific applications
• API integrations and workflows
• User interfaces for scientific computing

🌟 Open Innovation

• Creative applications of LLMs in science
• Cross-disciplinary approaches
• Novel use cases and methodologies

Judging Criteria

Projects will be evaluated based on:

Potential for Impact (25%)

Research advancement potential

Innovativeness (25%)

Unique approaches and creativity

Scalability (25%)

Growth and user accommodation capability

Relevance (25%)

Advancement of materials science and chemistry

Technical Guidelines

Recommended Technologies

• LLM APIs: OpenAI GPT, Anthropic Claude, Google Gemini, etc.
• Open Source Models: Llama, Mistral, CodeLlama via HuggingFace
• Scientific Libraries: RDKit, ASE, Pymatgen, PySCF, etc.
• Frameworks: LangChain, LangGraph, AutoGen, CrewAI

Data and Ethics

• Use publicly available datasets or generate synthetic data
• Respect copyright and licensing of data sources
• Include proper citations for datasets and models used
• Avoid sensitive or proprietary information
• Follow ethical AI practices in development and deployment

Code Quality

• Write clean, readable code with appropriate comments
• Follow language-specific style guides (PEP 8 for Python, etc.)
• Include error handling and input validation
• Provide clear setup instructions and dependency management

Resources for Participants

Computing Resources

• Cloud computing credits may be available (check with organizers)
• Local computational resources at on-site locations
• Free tiers of major cloud providers (AWS, GCP, Azure)

Datasets

Visit our Resources page for curated datasets in:

• Materials science databases
• Chemical reaction databases
• Scientific literature and publications
• Molecular and crystal structure data

Getting Help

• Slack community: Join here
• Mentorship sessions: Available throughout the event
• Technical support: On-site and virtual assistance
• Documentation: Comprehensive guides and tutorials

Submission Checklist

Before submitting, ensure you have:

•   Public GitHub repository with complete code
•   README.md with all required sections
•   Working installation instructions tested on clean environment
•   Demo video (2-3 minutes) uploaded and accessible
•   Example outputs or screenshots included
•   License file added to repository
•   Team member information documented
•   Submission form completed with all details

Questions?

For submission-related questions:

• Check our FAQ page
• Join our Slack community
• Contact organizers: blaiszik@uchicago.edu

Good luck with your submissions! We’re excited to see your innovative applications of LLMs in materials science and chemistry.