RationAI — PDS System at Human Resolution

RationAI

Inspiration

800 million Indians depend on the Public Distribution System. Most of them cannot answer three basic questions: What am I entitled to? When will it arrive? What do I do when it doesn't?

The information exists. It lives inside government portals built for administrators — not for the woman in rural Gujarat who needs her ration card linked, or the elderly man who hasn't received his quota in three months and doesn't know where to complain.

We asked one question: what happens when you point AI at that problem?

What it does

RationAI is a bilingual (Gujarati + Hindi) AI-powered navigator for India's PDS system. It does two things:

Optimise: A conversational AI guide that helps citizens:

Check their monthly ration entitlements
Locate their nearest fair price shop
File complaints in plain language
Understand their rights under NFSA

Represent: A live dashboard that makes the system legible — which districts report consistent shortfalls, complaint patterns by region, distribution gaps by commodity. The most powerful tool isn't always a fix. Sometimes it's a mirror.

How we built it

Built using Lovable (AI-powered frontend builder) with Claude AI as the conversational and research engine. We mapped the PDS data flow — central allocation → state distribution → fair price shops — and identified the 6 key friction points citizens face daily.

AI was embedded in every stage: problem research, system mapping, design decisions, content generation, and build. Not as decoration — as the actual method.

Challenges we ran into

PDS data is fragmented across 28 state portals with no unified API
Translating bureaucratic language into plain Gujarati accurately without losing meaning
Representing systemic failure without misrepresenting individual fair price shops
Building a working product in 24 hours with zero prior coding experience

Accomplishments that we're proud of

Working bilingual AI assistant live on the internet in under 6 hours
Zero coding experience → fully deployed web app
Dashboard visualising real Gujarat district-level PDS patterns
Identified the 6 most common PDS failure points citizens face

What we learned

The hardest part was not building the tool. It was understanding the system well enough to intervene usefully.

Government systems are not complex because complexity is necessary. They are complex because nobody ever designed them for the person at the end of the chain — the person who walks 4 kilometres to a fair price shop and is turned away because the system says their card is inactive.

AI made it possible to go from zero to working product in hours. But the decisions that mattered — which system to address, whose pain to centre, what "better" actually means for a rural citizen — those required human judgment.

We are humans inside the system of building. AI was the tool. Understanding was the work.