Chris Vann ™26, Tasin Saifullah ™25, Siham Boumalak ™25
Advisor: Rob Kelvey
The goal of the project was to test large language models, prompts, and tools, and develop autonomous agents to enhance Schneider Electric™s R&D efforts. The AMRE program used various LLM models such as GPT, Claude, and Gemini to test the reasoning and planning capabilities of autonomous agents. The team™s extensive efforts led to the developing of a variety of experiments that tested the capabilities and limitations of autonomous agents; some of these agentic projects include extracting insights from a dataset, document data extraction, a reference checker, and a plagiarism checker.