Logo
← Back to archive

AI Student Advisor

Archived

RAG chatbot answering Purdue course and degree questions, plus a graph planner that finds the shortest path to graduation.

Project Manager
Andrew Qiang

NLP Generative AI Software Engineering
Ran
Fall 2025 – Spring 2026
Commitment
4-8 hrs/week
Openings
0-1

What you need coming in

LLM techniques (various flavors of RAG) and NLP.

This project aims to create an LLM chatbot advisor based on official Purdue information.

Objectives

  1. Train a valid local LLM that can answer basic user queries without relying on RAG, in a format aligned with human preferences
  2. Optimize RAG pipeline speed, and lower RAG costs where possible
  3. Optimize the web UI and make the site more professional
  4. Train a graph tool for course planning that picks the shortest path to graduation given a student’s current classes