AI Agents for Engineers

24-880 • Spring 2026 • Carnegie Mellon University

course_logo.png

This course introduces graduate students and advanced undergraduates in engineering to the theory and practice of large language models (LLMs), with a focus on their application in engineering domains. Students will learn how to harness foundational models such as GPT, LLaMA, and Claude to solve real-world engineering problems, automate workflows, and build intelligent agents capable of reasoning, planning, and interacting with scientific tools and environments.

The course covers the fundamentals of LLM architectures, fine-tuning and prompting strategies, and tool-augmented LLMs, including retrieval-augmented generation (RAG), function calling, and agent frameworks. Emphasis is placed on designing autonomous agents that can assist in engineering tasks such as data analysis, simulation control, documentation generation, code synthesis, and system optimization.

Through hands-on projects and case studies, students will apply LLMs to domains including mechanical design, materials discovery, systems engineering, and computational biology. The course balances technical depth with practical implementation, preparing students to both understand the capabilities and limitations of LLMs and to build domain-specific, AI-driven systems that enhance engineering innovation.

Suggested Prerequisites:

Students should have experience with one of the following:

  • (24-787) Artificial Intelligence and Machine Learning (or)
  • (10-701) Introduction to Machine Learning (or)
  • (11-785) Introduction to Deep Learning

Or equivalent coursework/experience in:

  • Programming (Python proficiency required)
  • Basic linear algebra and probability theory
  • Familiarity with neural networks and machine learning concepts