---
title: "Syracuse Case Study, Expanded: Clementine in Production"
slug: "syracuse-case-study-clementine-production-data"
date: "2026-08-07"
tag: "Application"
summary: "The Syracuse University case study now carries the university's own published numbers — 2,038 students and 8,217 questions in 37 days of registration season — plus the classroom use cases Syracuse runs on the platform, and direct links to the four pages ITS publishes at ai.syracuse.edu."
author: "ibl.ai Engineering"
linkedin: |
  Syracuse University publishes its own AI program at ai.syracuse.edu — and lists Clementine, built on ibl.ai, as one of four University-approved AI platforms alongside Claude Enterprise, Copilot, and Gemini.

  They also published the usage data. We've now folded it into our case study, in their numbers, with links back to their pages:

  → 2,038 students asked 8,217 questions across 2,401 conversations in 37 days of registration season
  → 41% of those questions landed inside the 5-day registration window
  → 37% of usage happened after hours
  → Two-thirds of conversations were multi-turn — students going back and forth, not one-shot lookups

  What we find most interesting isn't the volume. It's what a university does once it owns the thing.

  Syracuse's ITS team builds course agents grounded in a professor's own material — a "virtual Jeff" that answers from the actual syllabus. Generated practice exams. Branching simulations. Student study agents. TAs scaled across sections. Agents embedded in other systems through the API, with grounding sources and guardrails set per agent.

  And because the platform is model-agnostic, Syracuse authors pick Claude, ChatGPT, or Gemini per agent — and switch anytime. That's not a feature we shipped for them. It's what the deployment model makes structurally possible: you own all the code and the data, so nobody else decides which model you're allowed to run.

  Read Syracuse's own write-up of why they bought the source code instead of renting a platform: https://ai.syracuse.edu/stories/ibl-google-cloud.html

  Our expanded case study: https://ibl.ai/case-study/syracuse-university

  #iblai #HigherEd #AgenticAI #EnterpriseAI #AIinEducation
---

Syracuse University's ITS organization publishes its own AI program at [ai.syracuse.edu](https://ai.syracuse.edu/), where **Clementine** — the university's AI platform, built on ibl.ai — is listed as one of four University-approved AI platforms available to every student, faculty, and staff member, alongside Claude Enterprise, Copilot, and Gemini.

Three of those pages carried material our case study was missing: first-party usage data from registration season, and the classroom use cases Syracuse actually runs. [The case study](https://ibl.ai/case-study/syracuse-university) has been extended with all of it, attributed and linked back to Syracuse throughout.

## In Production — Clementine Class Search

Syracuse published a usage report from the Fall 2026 registration window. The numbers below are theirs, not ours:

| metric | Syracuse's published figure |
|---|---|
| students | 2,038 |
| questions | 8,217 |
| conversations | 2,401 |
| window | 37 days |
| inside the 5-day registration window | 41% of all questions |
| after hours | 37% of usage |
| multi-turn conversations | roughly two-thirds |

The case study now renders these alongside verbatim student queries from Syracuse's report. Two-thirds of conversations being multi-turn is the detail worth sitting with — students were working a problem, not running a lookup.

## Use Cases

A new section covers what Syracuse's faculty and ITS team build on the platform: the "virtual Jeff" course agent grounded in a professor's own material, generated practice exams, branching simulations, student study agents, scaling TAs across sections, and agents embedded in other systems through the API — each with its own grounding sources and guardrails.

It closes on model-agnostic proof: Syracuse authors choose Claude, ChatGPT, or Gemini per agent, and switch anytime. That is what owning the code makes possible, rather than something the vendor grants.

## In Syracuse's Own Words

The case study now links directly to the four pages Syracuse publishes:

- [AI Tools at Syracuse](https://ai.syracuse.edu/ai-tools.html) — the four University-approved AI platforms available campus-wide, Clementine among them.
- [Owning Its AI: Syracuse, ibl.ai and Google Cloud](https://ai.syracuse.edu/stories/ibl-google-cloud.html) — why Syracuse bought the source code instead of renting a platform, how it runs on the university's own Google Cloud, and what 85% lower cost looks like in practice.
- [Clementine — Syracuse's own AI platform](https://ai.syracuse.edu/clementine.html) — the campus-facing documentation: what you can build, how to ground an agent in your material, choosing the model, setting guardrails, controlling access.
- [Clementine Takes On Registration Season](https://ai.syracuse.edu/stories/clementine.html) — the usage report the numbers above come from.

Also corrected in the same pass: Jeff Rubin's title is Senior Vice President & Chief Digital Officer, and his testimonial — previously missing from the page's markdown counterpart — is now carried there too.