Skip to content
SIH Buddyby Ganeev Singh
Dev
All problem statements
SIH26092Proceed with cautionacceptance 2/5

AI-Driven Scheme Matching for Marginalized Entrepreneurs

Ministry of Social Justice and Empowerment (MoSJE) · Miscellaneous · Software

A rule engine and a map dressed as an AI recommender, overlapping heavily with the statement beside it — the one feature that would give it substance depends on partner financial data nobody will give you.

What it actually is

Concessional loans for this group are not given directly but routed through more than a hundred channel partners — state agencies, banks, rural banks and microfinance companies — and an applicant has no way to know which scheme fits them or which partner near them handles it. Applications go to the wrong place and stall. The ask is a platform that matches the person to the scheme and to the right nearby partner.

What to build

A three-part platform: a recommender taking project type, estimated cost, income and education status and returning the applicable scheme by applying the published eligibility rules rather than by inference, since eligibility here is a deterministic rule set and a learned model would be both unnecessary and unaccountable; a calculator producing projected repayment across the interest rate and moratorium ranges the statement gives; and a partner locator mapping the applicant to the nearest channel partner authorised for that specific loan category, with the harder requirement the statement adds of filtering by whether the partner is currently in a position to lend at all.

Smallest thing that wins the room

Enter an applicant profile and watch the correct scheme resolve with the eligibility rule that decided it shown explicitly, then see the two nearest partners authorised for that category appear on the map with the ones excluded and the reason why.

How crowded this one gets

A guess, projected from the 2025 statements — the last year where both the submission counts and the winners were published.

Moderate160–360 teams expectedroughly 1 in 132–305 wins it

Quieter than 24% of the 226 · #171 of 226 by expected field

A normal-sized field. Your idea has to be good, not miraculous.

Why: central ministry statements sat below the average.

This is a guess, not a fact

Nobody has published 2026’s numbers yet. This is an analysed estimate from last year’s pattern, so please do not take it as the truth — check the live counter on the SIH portal before you decide anything. The range covers the middle half of likely outcomes, so one statement in two lands outside it. Entry closes at 500 ideas per statement, so no range goes past that — a statement that reaches the cap fills and shuts rather than drawing an unlimited crowd. The model reads only three things a team can see before choosing — software or hardware, the theme, and what kind of body posted it — and those explain about a quarter of the variation in last year’s field sizes (R² 0.25 on held-out statements). Trust the band more than the number, and the ordering more than either. It cannot see how good your idea is, which is the part that actually decides it.

The scores

The number is the shorthand. The line under it is the reason.

What you will be writing

  • deterministic eligibility rule engine
  • geospatial nearest-partner matching with category filter
  • amortisation across variable moratorium periods
  • channel partner directory assembly
  • multilingual eligibility explanation
  • PostGIS partner coverage mapping
  • Concessional credit access
  • Financial inclusion
  • Government scheme delivery

Prior art to read before you start

scheme eligibility matching and recommendation · geospatial service provider routing · loan repayment calculation

Analysed by Claude Opus. Every score above is a judgment call with its reasoning attached — kindly cross-check this against the official statement on the SIH portal before your team commits to it.