Build agents
Topics: LLM agents, tool use, planning, memory, retrieval, code, multi-agent orchestration.
CS 839 · Fall 2026
AI Agents for Science
Our goal: to study the foundations and frontiers of AI agents, with an emphasis on building reliable systems for scientific discovery.
Course at a glance
Do science
Topics: Literature synthesis, hypothesis generation, experiment design, simulation, data analysis, automated labs.
Make agents trustworthy
Topics: Evaluation, reproducibility, verification, safety, monitoring, responsible deployment.
Logistics
- Term: Fall 2026
- Time: Tues. & Thurs., 9:30–10:45 AM
- Location: Mechanical Engineering 1164
- Instructor: Frederic Sala
- Office hours: Tuesday 3:30-5:00 pm, Morgridge 5514
- Course communication: Slack (link TBD)
Course site in progress. The syllabus and schedule are preliminary and will be updated. Check back in often!
Who should take this course?
This is an advanced, research-oriented course predominantly aimed at students interested in AI for science, advanced AI, machine learning, scientific computing, or computational science. Familiarity with core machine learning concepts and comfort reading research papers are expected.