Home › Hard problems › India

India: hard problems at India's scale

Addresses that machines can't read

Solo: strongOne person can run a real test of this in about 8 weeks.

the address, as people write itGround floor, behindthe temple, near theold well, Ward 7Map softwareturns text intoone map pointa templeand a wellanothertemple,anotherwellthe real doorthe pin lands herethe rider, the ambulance and the bank'saddress check all go to the pinDIGIPIN, a national digital addresscode, is new and not yet on most
Indian addresses are directions in free text, not codes, so the map point lands at the wrong temple and deliveries, ambulances and KYC checks miss.

Why it is still open

Indian addresses are free text ("behind the temple, near the old well"). Deliveries, ambulances and KYC fail on them. DIGIPIN, a national digital address code, is new.

An experiment one person can run in 8 weeks

Parse and geocode 5,000 public-style addresses with 3 methods and measure how many land within 100 metres.

Verdict: Solo: strong

Clear numbers, and logistics and fintech care.

The solo test behind the verdict

  1. Can it be measured? There is a number that shows you did better, such as an error rate, a catch rate or a cost per answer.
  2. Is it cheap to try? It runs on a laptop, free open models and open data, or under about ₹50,000 of rented GPU time.
  3. Is there a public yardstick? A benchmark or dataset others use, so your result can be compared and believed.
  4. Would a small win get noticed? A lab, a funder or a customer would care about a 10% improvement, not only a full solution.

Four yeses means Solo: strong. Two or three means Solo: possible. Fewer means Follow, don't build.

PreviousUseful AI on a ₹10,000 phone, offlineNextVoice-clone and UPI scams in Indian languages

All 79 hard problems · Working on this, or know of a result we missed? basant@cleandataengineering.ai