Scaling Advising: A Multi-Agent RAG System with Institutional Guardrails for Student Success

Thursday, October 01, 2026 | 9:30AM–10:30AM MT
Session Type: Poster Session
Delivery Format: Poster Session
What does it take to safely use large language models (LLMs) for academic advising? This session shows how a multi-agent, guardrail-driven AI assistant can deliver accurate, citation-backed answers to complex policy questions, reduce advising bottlenecks, and maintain trust by deferring high-risk requests to human advisors.

Presenters

  • Austin Blackwell

    Artificial Intelligence Developer & Systems Architect / Adjunct Professor, University of South Carolina

  • Elizabeth Shirkey

    Executive Director of IT Service and Customer Engagement, University of South Carolina