Chetu Builds an Agentic AI Student Assistant Platform for a Public University

The Client

Our client is a public university in North Carolina, serving roughly 3,000 students through the state's tuition assistance program. Its student population includes a high share of first-generation students, student athletes, and veterans, each with distinct support needs. Prior to engagement with Chetu, student support was spread across separate offices — Registrar, Financial Aid, IT Helpdesk, Student Success, and athletics; each on its own hours.

Unfortunately, its support model operated reactively and required rethinking. Fragmented offices did not reflect the experience the client wanted, and no single office could answer a question spanning more than one system. The university sought a unified, proactive agentic AI assistant that could securely connect information across departments and support students without expanding staff headcount.

The Challenge

There was extreme disjunction between the various offices and systems, obstructing information access and hindering cross-channel support. A single student question could touch four or five systems at once, and no office had visibility into all of them together.

The strongest risk signals—a registration change, a shift in academic standing, or a loss of satisfactory academic progress for financial aid—surfaced only after a student had already begun to disengage.

The ultimate goal was to unify these channels through an artificial intelligence-powered conversational interface, enabling every student to obtain accurate, personalized information around the clock without compromising FERPA compliance. The platform also needed AI agents capable of securely navigating multiple campus systems and coordinating responses across academic, financial, wellness, and student-life functions.

The Solution

This venture is ongoing, existing as a joint effort with the university’s shared services organization to expand agentic AI functionality that fits the client’s operations. Chetu’s Data Science team designed the platform and produced a formal, field-by-field data specification, which was validated in joint working sessions before development began.

Technologies Implemented: A combination of Custom GPTs, an MCP connector layer, Ellucian Banner, a vector database for unstructured content, and Microsoft Graph API were utilized in this project.

The Process: In order to fix a fragmented student support experience, Chetu developers:

The Impact

The agentic AI platform creates a more cohesive and proactive student-support experience. Registration and academic-standing changes that previously surfaced only after the fact can now prompt earlier outreach and more timely intervention.

Chetu implemented specialized AI agents that make academic, financial, and wellness information easier for students to access while handling repetitive tier-one questions that previously consumed staff time. This helps reduce response times, improve support efficiency, and allow staff to focus on more complex student needs.

By using artificial intelligence to unify siloed support functions and replacing ad hoc data sharing with tiered, documented access, the platform enables students and advisors to connect more efficiently. As adoption grows, the client can expect higher student satisfaction, stronger engagement with support services, and improved student-success outcomes.