I Built a CET Counsellor AI Agent for Maharashtra Engineering Aspirants — Sanity Challenge

I developed an AI assistant called CET Counsellor to assist Maharashtra engineering aspirants with navigating the complex admissions process. The system addresses the confusion surrounding the CAP admission rounds, 13 category-wise cutoffs, required documents, and more for MHT-CET and JEE exams. The live website is accessible at https://cet-guide-ai.lovable.app and the Sanity project ID is op01iqirw.

The CAP admission process is notoriously confusing for students, with multiple rounds, 13 seat categories, varying documents, and shifting cutoffs each year. Having gone through the process myself this year, I understand the stress it creates.

The knowledge base for CET Counsellor is structured in Sanity, consisting of 5 CAP stages explained step-by-step, 6 colleges with branch-wise closing percentiles, and 13 categories including OPEN, OBC, SC, ST, VJNT, EWS, TFWS, PWD, Defence, Ex-servicemen. Each category has its own reservation rules and required certificates. Sanity's structured content allows the AI agent to query the information meaningfully, rather than merely matching keywords.

Using React for the frontend and TanStack CMS integrated with Sanity (Project ID: op01iqirw, dataset: production), the site was deployed via Lovable. Building this project as a first-year Computer Science Engineering student on a mobile device, without a laptop, reinforced the fact that the idea behind the solution matters more than the specific tools used. Sanity's structured approach enabled the creation of a complex admission data system that actually helps students.