This is a submission for the Sanity Challenge. Path One: Ship an Agent That Queries Real Content
What I Built
CET Counsellor is an AI agent that helps Maharashtra engineering students navigate the MHT-CET and JEE CAP admission process. Students can ask about CAP rounds, category-wise cutoffs, required documents for all 13 seat categories (OPEN, OBC, SC, ST, VJNT, EWS, TFWS, PWD, NT-B, NT-C, NT-D, Defence, Ex-servicemen) in English, Marathi or Hinglish.
The Story
I went through CAP admissions myself this year
(MHT-CET 98.46 percentile, VIT Pune CSE).
The process is genuinely confusing. So I built
the tool I wish I had — entirely on mobile,
no laptop, no terminal.
Demo
🔗 Live site: https://cet-guide-ai.lovable.app
Code
Built with Lovable (React + TanStack) connected to Sanity as the knowledge base CMS.
🔗 GitHub: https://github.com/05unique-dotcom/CET-Counseller
How I Used Sanity
Sanity stores the complete structured knowledge base:
•201 colleges across Pune and Mumbai region
•555+ branches — Computer Engineering, IT, AI & Data Science, ML, Cyber Security
•CAP Round 1, 2 & 3 closing percentiles sourced directly from official Maharashtra CET Cell PDFs
•13 seat categories — OPEN, OBC, SC, ST, VJNT, EWS, TFWS, PWD, NT-B, NT-C, NT-D, Defence, Ex-servicemen with reservation rules and required documents
The AI agent queries this structured Sanity content to give accurate, meaningful answers — not just keyword matches. A keyword search would never tell you "MIT Alandi is a close call at your percentile" — but structured data can.
Sanity Project Details
- Project ID: o1nu2vvm
- Dataset: production
Agent Session
No transcript available.
Top comments (0)