Scholarship Intelligence

Inspiration

The National Scholarship Portal, Buddy4Study, and tools like CheckMyEligibility already help students discover scholarships. But in our pilot survey of 38 students, 61% had found a scholarship but were unsure whether they actually qualified, and most reported eligibility criteria and missing documents as their biggest frustration. Discovery is a solved problem. Knowing whether you truly qualify is not.

What it does

Scholarship Clarity takes a student's profile and checks it against a scholarship's actual eligibility rules, one rule at a time. Every rule returns PASS, FAIL, or UNKNOWN. Missing evidence never becomes a guessed "yes." The student sees exactly which rule passed, which failed, what evidence is still needed, and what to do next, ranked by deadline urgency, benefit, and how ready their documents are.

How we built it

We catalogued 16 real scholarships (13 from NSP, 3 from other sources) into a rule matrix, with each rule tagged PASS/FAIL/UNKNOWN logic, a source link, and a verification date. We tested four existing platforms (NSP, Buddy4Study, CheckMyEligibility, and Tamil Nadu's welfare department site) against the same student profile to map what they already do well, and where students are still left guessing. Our proposed architecture uses AI to read official scholarship text and propose rules, which a person verifies; the final eligibility decision always comes from an auditable rule engine, not the AI.

Challenges we ran into

Our own survey data had a real trap: pie-chart colors don't always map to answers in the order you'd expect, and we caught and fixed more than one misread stat before finalizing our numbers. Scholarship rules also change often and sometimes conflict across sources (we found two different income caps for the same Tamil Nadu scheme), which shaped our decision to store a source and verification date on every single rule.

Accomplishments that we're proud of

Testing our concept against real platforms with a real student profile, instead of assuming a gap existed. Catching and correcting our own survey and testing errors before submission, rather than presenting numbers we couldn't defend. Landing on a genuinely different value proposition: reasoning over a decision, not just matching.

What we learned

The interesting problem was never "which scholarships exist." It's "can I prove I qualify, and what am I still missing." Explainability and honest uncertainty (UNKNOWN, not a guessed yes) matter more than adding another recommendation algorithm.

What's next for Scholarship Clarity

Round 2: a working prototype, tested against a manual baseline for time-to-decision and eligibility accuracy, starting with Tamil Nadu students and expanding to more scholarship sources and states.