Live in-pipe leak inspection of large water mains, paid per kilometre
TL;DR Linesong checks the big pipes that feed Indian cities from the inside, while water keeps flowing through them. An acoustic tool goes in through a tap, rides along with the flow and records every leak and air pocket it passes. City water departments pay per kilometre surveyed and per leak they dig up and confirm.
Indian cities lose a scary amount of treated water before it ever reaches a meter. The government's expert committee put non-revenue water at about 50% of production in 2011, and a parliamentary committee said in December 2025 that cities had barely moved despite incentives to get below 20%[1].
The biggest pipes are the hardest to check. Mumbai runs mains from 50 mm up to 2,200 mm, and anything above 300 mm sits 7 to 20 feet down, usually under a concrete road. So in 2026 the city's hydraulic engineers asked firms to pitch non-destructive ways of finding leaks on these lines, because digging just to look is slow and costly[2].
Can you blame them?
The seed pictured AI sensors built permanently into city pipes, and I've reframed it. Most of a network is small pipe with no power or signal underground, and nobody is going to pay to fit every metre. Large mains are a different story - there are few of them, they're expensive to dig and every leak loses a lot of water.
So Linesong runs survey campaigns on those mains. A crew fits a launch fitting to an existing tap, releases a ball-shaped acoustic recorder and tracks it from the surface as it travels. At the far end the tool is caught and its data is read. The report marks leaks, air pockets and the pipe's path to within a few metres, ranked by likely loss.
This has been done abroad for years. One global water-tech firm's free-swimming leak tool has been in use since 2007, goes in through a 100 mm tap on a live line, and in one Canadian survey of a 1,220 mm main it placed a leak within a metre of where crews later dug[3]. Linesong starts by licensing or partnering on tools like that, and builds its own only if the volume is there.
The water department or its operator pays. That means municipal corporations, state water boards and private firms that run city networks on performance contracts. Leak work is already bought this way. Delhi Jal Board asked for bids on a performance-based leakage contract in 2016[4], and Kolkata gave a global water operator a six-year €30 million (about ₹325 Cr) contract to cut losses in one area[5]. Linesong would charge per kilometre plus a fee per confirmed leak. I couldn't find an Indian price list for this work, so that's yours to test.
Cities are actually asking for this now. Mumbai's 2026 call for non-destructive leak technology on mains up to 2,200 mm is the clearest sign[2]. The parliamentary review of AMRUT in December 2025 also pushed cities on their water losses[1], and funding is more and more tied to measured results.
There is no public figure for in-pipe inspection of water mains in India. What you do have is buyer behaviour - Mumbai's open call[2], Delhi's performance-based leak tender[4] and Kolkata's €30 million loss-reduction contract[5]. Treat the 50% loss figure[1] as background for the whole network, not as the size of this market.
That global water-tech firm sells free-swimming inspections worldwide[3]. Big international water operators take on whole-network loss contracts, as in Kolkata[5], and leak detection is just one tool in their kit. Indian leak-survey firms mostly listen from the surface. I found no Indian company whose core business is in-pipe inspection of large water mains, and that's your gap.
| Difficulty | Hard (54/100) |
| Starting capital | ₹1.2–2.4 Cr |
| Time to launch | 9–15 months |
If you partner for the tool then the tool isn't your problem. Field work and selling are. Your crews will work on live mains at night alongside the utility's staff, and every sale crawls through tenders and committees before anyone signs. Expect to wait several months to get paid.
No licence is needed to inspect pipes, but procurement, radio and water hygiene rules apply.
Building this outside India? The categories above usually have a local equivalent, so check with your own regulator.
Don't trust AI alone to place a leak for digging. An analyst checks every flagged point, because a wrong dig on a main under a concrete road is expensive.
Want to work out the AI layer for Linesong? Talk to Yash →
This suits a founder from water utilities, pipeline inspection or a large water operator, someone who doesn't mind standing next to an executive engineer at 2 a.m. If you're hoping to sell software then it is not for you. I'd first prove that one city will pay for a 10 km pilot and dig at least three of the leaks you flag.
Plan for roughly ₹1.2–2.4 Cr, mostly on tool access, crews and the cash to survive slow municipal payments.
No sector licence. You need WPC approval for any radio tracker, each utility's hygiene approval and the documents tenders ask for.
Yes, water loss is a global problem and the tools are already used abroad. Check how your local utilities buy leak work and their hygiene rules.
The free Linesong validation kit has the line-by-line capital breakdown, the effort scoring, the procurement and radio checklist, a 14-day validation sprint with pass/fail targets, a 90-day launch plan, 10 interview questions for utility engineers, and 5 assumptions to test. Download the kit →
Thinking about building Linesong or something close to it? The make-or-break bit is locking in a tool partner and getting one city to dig where your survey points, and that's the kind of thing I help founders work through. If you need a hand, you know where to find me. Talk to Yash →
For illustration only. Names are fictional and ideas hypothetical; sourced facts are real but may be out of date. Nothing here is legal, financial or professional advice, and Yash Khatri is not personally liable for any use of it. Read the full disclaimer.