Product Name: orgGPT

Tagline: “Every question, one answer away.”

Applying to programs is a common step but an exhausting hassle, no student can deny it. Try looking for: what’s the average GMAT score for Cornell’s MPS in Management? If you can find the right answer “640-660” online, reject my idea now. (The answer is provided by the admissions committee.)

Core Problem: Students face challenges finding specific university information, while universities are overwhelmed with repetitive queries.

Core Solution: AI-based, university-specific GPT platform that centralizes and answers student/staff queries instantly, leveraging institutional data.

Primary Customers:

• University administration teams, including admissions, academic advising, and student services.

End Users:

• Students: For academic, event, and campus service inquiries.

• Prospective Students: For program and application related information.

• Faculty and Staff: For procedural and administrative guidance.

• Alumni and Visitors: For event and networking information.

Value Proposition:

• For Universities: Reduces repetitive inquiries, allowing staff to focus on strategic tasks, offers analytics to improve operations, and enhances reputation via better student engagement.

• For Students: Provides immediate, 24/7, official and reliable answers without the need to navigate multiple websites.

---

Solution Approach:

I have been taking guidance from Professor Emaad Manzoor (who teaches AI for Business Applications at Cornell) to refine the technical approach, concluding that Retrieval-Augmented Generation (RAG) training is the best fit, with data collection as the secret sauce. Specifically focusing on university email data to keep the information current. Along with the current documentations and FAQs, the system will automatically monitor staff emails, with an option for privacy (allowing staff to disable monitoring for sensitive emails). This filtered data would then be fed into the model.