SpaceX AI Assistant - Hackathon Project

Overview

Welcome to the repository of our LLM (Language Learning Model) project, developed for a hackathon competition. Our model is built on an open-source LLM and is trained using NASA data. The primary objective of this project is to enhance the efficiency of private space companies like SpaceX by providing an AI-powered semantic search assistant.

Objective

Astronauts currently spend 1-2 hours daily searching through paper manuals and PDFs to locate critical information. Our AI platform is designed to act as a semantic search assistant, integrated with Mission Control, to streamline various processes including communication, health monitoring, and task management. By saving an estimated 2 hours per astronaut per day at a cost of $130,000 per astronaut hour, this solution offers substantial savings for commercial space station operators.

Features

• Semantic Search: Advanced AI-powered search capabilities to quickly find relevant information from extensive documentation.
• Contextual Understanding: The AI model provides contextual answers to complex queries, similar to Perplexity AI, enabling astronauts to get precise and relevant information.
• Interactive Q&A: The model can engage in interactive question-and-answer sessions, providing astronauts with detailed explanations and insights based on NASA data.
• Documentation Summarization: Summarizes lengthy documents and manuals into concise, easy-to-understand formats, saving time and improving comprehension.

Contributing

We welcome contributions to enhance the functionality and performance of the SpaceX AI Assistant. Please fork the repository and submit pull requests with detailed descriptions of your changes.

License

Some contents of this repository are based on open-source models, and their usage is subject to their respective licenses.

Acknowledgments and Credits

We would like to thank the open-source community and NASA for providing the data and tools necessary for this project. Special thanks to the hackathon organizers for the opportunity to work on this exciting project.

PrivateGPT credits: Martínez Toro, I., Gallego Vico, D., & Orgaz, P. (2023). PrivateGPT [Computer software]. https://github.com/imartinez/privateGPT

For any questions or support, please open an issue in this repository or contact the project maintainers.

Happy coding! 🚀