Skip to content

rushikd27/llmhackathon.github.io

 
 

Repository files navigation

LLM Hackathon for Applications in Materials Science & Chemistry

This is the official website for the LLM Hackathon for Applications in Materials Science & Chemistry, a hybrid international hackathon connecting researchers from across the globe to explore and solve problems in materials science and chemistry using Large Language Models.

About the Hackathon

The LLM Hackathon brings together the brightest minds from academia and industry for a weekend dedicated to solving critical challenges in materials science and chemistry using the power of Large Language Models. This event showcases innovative applications at the intersection of AI and molecular science.

2025 Event Details

Dates: September 11-13, 2025

  • Kick-off: September 11, 9:30 AM CST
  • Hacking Period: September 11-12
  • Submission Deadline: September 13, 9:59 AM CST
  • Awards Ceremony: September 19, 9:30 AM CST

Format: Hybrid (online + on-site locations)

On-Site Locations

  • London: King's College London
  • New York: Cornell Tech (Roosevelt Island)
  • Tokyo: University of Tokyo

Previous Hackathon Results

2024 Hackathon

34 projects were submitted across 7 on-site locations and virtual participation.

Publication: Reflections from the 2024 Large Language Model (LLM) Hackathon for Applications in Materials Science and Chemistry - arXiv Preprint

Selected Projects Publication: 34 Examples of LLM Applications in Materials Science and Chemistry: Towards Automation, Assistants, Agents, and Accelerated Scientific Discovery - arXiv Preprint

Prize Winners:

  • 1st: LangSim
  • 2nd: GlossaGen
  • 3rd: PoreVoyant
  • 4th: Team Datalab

Additional prizes awarded by Neo4j, Anthropic, and Reincarnate.

2023 Foundation Hackathon

14 pioneering projects established the foundation for LLM applications in materials science and chemistry.

Publication: 14 examples of how LLMs can transform materials science and chemistry: a reflection on a large language model hackathon - Digital Discovery

Key Resources Featured

The website provides comprehensive resources including:

Learning Materials

  • LLM fundamentals and architecture tutorials
  • Library-specific guides (RDKit, PySCF, ASE, Pymatgen, LangChain, LangGraph)
  • Fine-tuning and RAG technique tutorials

Datasets

  • Materials Science: Materials Project, NOMAD Laboratory, Crystallography Open Database
  • Chemistry: PubChem, ChEMBL, Open Reaction Database, USPTO Reaction Datasets
  • General: arXiv preprints, Hugging Face datasets, Kaggle collections

Research Papers

  • Key reviews on LLMs in materials and chemistry
  • Survey papers on AI for materials science
  • Foundation models and autonomous agents research

Prizes & Recognition

Prize Structure

  • 1st Place: $1,000
  • 2nd Place: $500
  • 3rd Place: $250
  • Special Awards: Various sponsor prizes

Judging Criteria

  1. Potential for Impact: Research advancement potential
  2. Innovativeness: Unique approaches and creativity
  3. Scalability: Growth and user accommodation capability
  4. Relevance: Advancement of materials science and chemistry

Organizing Team

Lead Organizer: Dr. Benjamin J. Blaiszik
University of Chicago and Argonne National Laboratory

Volunteers: Adib Bazgir, Alexander Al-Feghali, Aritra Roy, Hassan Harb, Piyush R. Maharana, Pranav Krishnan, Thomas Pruyn

Sponsors & Partners

  • LILA (Learning in Artificial Intelligence and Applications)
  • HuggingFace

Important Links

Website Features

The website includes:

  • Event information and schedule
  • Comprehensive participant resources
  • Past project showcases with links to repositories
  • Submission guidelines and FAQ
  • Information about on-site locations and hosting requirements

This hackathon represents the collaborative effort to advance the intersection of artificial intelligence and molecular science through community-driven innovation.

About

Official website for Hackathon on LLM Applications for Materials and Chemistry

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages

  • HTML 89.7%
  • CSS 10.3%