Background:
The Knudsen Institute is a 501(c)(3) nonprofit applied research organization whose mission is to identify and develop technology, workforce, & policy solutions through applied research and outreach that allows the US manufacturing capacity of small and medium sized manufacturers to be identified, evaluated, and integrated into the US Defense Industrial Base as rapidly as possible in a surge environment and to do so at scale.
Project Description:
Free text is now data thanks to AI. In this project, students will first explore AI-based web-scraping methods to understand their advantages over conventional web-scraping methods. Building on this, students will design and refine prompts that extract supplier capacity, certifications, and supply chain risk factors from the unstructured scraped text. Finally, students will implement an agentic pipeline that automates the entire collection-and-analysis workflow, producing a working prototype suitable for further research or deployment. Students will contribute to an open-source project so they can showcase their contributions to everyone.
Keywords:
AI, AI Engineering, Agentic AI, Generative AI, LLM, NLG, text processing, web scraping, open-source
Tools/Skills that will be used/learned:
Python, PyTorch, Hugging Face, GitHub
- Fall mentor time: Thursday: 3:30 PM Eastern
- Fall lab time: Tuesday: 3:30 PM Eastern
- Tools: github, python, pytorch
- Topics: ai, llms, nlp, web scraping
- Requirements: Open to all students