Subscribe
Share
Share
Embed
Stories and research on grassroots social organizations around the world — real people, real impact.
[Host] Local Action. You don't find fellow travelers in slogans — you find them in problems. Hello everyone, welcome. I'm Lili.
[Guest] Hi, I'm Tong.
[Host] I'm your host today. Where we're headed today is Dhaka, the capital of Bangladesh. Before we get there, I want to ask a question first: in 2026, if I handed you a smartphone, what's the most ordinary, everyday thing you'd think to do with it?
[Guest] Scroll short videos, answer messages, pay for things, check a map — right? What you probably wouldn't guess is that in Bangladesh, someone is using a smartphone to do something completely different: asking a large language model — "How often does the arsenic filter module on this water treatment unit need replacing? After the last change, the arsenic level in the output water is still high — what do I do?"
[Host] A large language model answering water-treatment questions — that sounds like a tech company's promo reel. But behind it is a longer story: how did a nanotechnology developed in a lab become a national infrastructure of 300-plus water ATMs and 700-plus treatment plants, reaching three million people? This story starts with a pair of numbers — twenty million to forty-five million.
[Host] According to World Health Organization data, twenty to forty-five million people in Bangladesh drink groundwater with arsenic above the safe limit — what does that mean in scale? Bangladesh's total population is roughly 170 million, so nationwide, that's roughly one in every four or five people drinking water that contains arsenic.
[Guest] Arsenic contamination is a problem that rarely makes headlines but runs extremely deep. The World Health Organization has called it "the largest mass poisoning of a population in history." Chronic arsenic poisoning causes skin lesions, cancer, and irreversible damage to children's neurological development. The root of the problem is unusual: Bangladesh's groundwater aquifers naturally contain arsenic — it's not factory discharge, not agricultural runoff — it's the geology itself. So drilling a well and drawing on groundwater is, in effect, drinking arsenic.
[Host] So has no one tried to solve this?
[Guest] Actually, quite a few teams have tried. The traditional approaches mostly go two ways. One is reverse osmosis, RO — it works well, but the equipment and running costs are too high for rural communities, and RO wastes 40% to 60% of the water fed into it. The other is boiling, but boiling doesn't remove arsenic, and it burns through a lot of fuel — in Bangladesh, that means cutting down more trees or buying more coal.
[Host] So the technology exists, but in Bangladesh it's — too expensive, too power-hungry, too wasteful of water, or it just doesn't work.
[Guest] Right. That's the starting point for Drinkwell's three co-founders: Minhaj Chowdhury, Dr. Mike German, and Professor Arup K. SenGupta. Minhaj is a Bangladeshi American who learned how severe Bangladesh's arsenic problem was while studying public health at Johns Hopkins. Mike holds a PhD in environmental engineering and did his research in Professor SenGupta's lab at Lehigh University. That lab had developed HIX-Nano, a hybrid ion-exchange nanotechnology — it sounds complicated, but it comes down to three core traits: it removes arsenic and fluoride, it can be regenerated and reused, and it costs 70% less than reverse osmosis.
[Host] Three traits — removes it, reusable, 70% cheaper than RO. But technology is technology. Turning it from a lab discovery into infrastructure that three million people use every day — that gap is not a small one.
[Guest] You've put your finger on it. Minhaj and Mike had the idea in 2012 and formally founded the company in 2013. Over the next few years they gradually commercialized the technology, and at first the approach was straightforward — sell filtration units, install them door to door.
[Host] And then?
[Guest] And then they ran into a fundamental problem. When they went to negotiate with local governments in Bangladesh, the answer was: "It's not that we don't need your technology — it's that we don't have the money to buy your equipment."
[Host] Even the utility companies didn't have the money?
[Guest] Utility companies in a lot of developing countries genuinely lack money, technology, and maintenance staff. Sell them an RO unit, and it might be sitting idle three months later — no one to repair it, no budget to replace the filter cartridges. Minhaj later summed up their thinking at the time in an interview: if all you're doing is selling equipment, your impact stops the day you make the sale. You have to find a way for your impact to grow together with the utility.
[Host] So how did they find that path? If not selling equipment, then what?
[Guest] The pivot in their thinking was the key move — don't compete with the utilities, partner with them. Not "I'll build an independent water system and you stay out of it," but "I'll embed myself in your system — the water is still yours, the pipes are still yours, and I'll run the hardest part for you."
[Host] What exactly is this "hardest part"?
[Guest] Removing arsenic and fluoride. Bangladesh's utility companies manage the pipe network, collect water fees, and handle basic maintenance, but arsenic contamination requires specialized technology. What Drinkwell does is: you handle the pipes, I handle the purification. They build water ATMs — a self-service kiosk in the community, like a vending machine — residents insert coins or pay by phone and get purified, safe water. Each water ATM connects to a utility's existing water point, and Drinkwell operates and maintains it, guaranteeing the water quality meets standard.
[Host] The model itself isn't hard to follow — but 300-plus water ATMs, 700-plus treatment plants — that's not something you build village by village on your own, is it?
[Guest] Of course not. What actually turned Drinkwell from a startup into a national infrastructure operator was something that happened in 2026 — the Asian Development Bank approved a 138 million dollar sovereign loan.
[Host] A 138 million dollar sovereign loan for a social enterprise? How does that even happen?
[Guest] You have to think backward: if Drinkwell were just a company selling water-treatment equipment, ADB would never lend to it. But Drinkwell's model is that they partner with every single utility company in Bangladesh — nationwide coverage. Once your operating network becomes national infrastructure, your customer is no longer just the end user — it's the country's water-security system. What this ADB loan really is: the Bangladeshi government using its sovereign credit as guarantee to lend Drinkwell money, so it can roll out more water ATMs and treatment plants nationwide. This isn't a commercial loan — it's a national-level infrastructure investment.
[Host] So... a lab technology walked a path from university lab to startup, then utility partner, and finally national infrastructure.
[Guest] Right. From the lab to national infrastructure — a long road, but one that worked.
[Host] Okay, one last question — Drinkwell has served three million people, 2.5 billion liters of water, 700 treatment plants. But Bangladesh still has twenty to forty-five million people drinking arsenic-contaminated water. Do you think that counts as success, or is it just the beginning?
[Guest] Both sides have a point. If you only look at the numbers — three million against twenty-plus million — the gap is still huge. But if you look at the path — a university lab technology, driven by partnering with utilities, financed by a sovereign loan — this path itself might be replicable.
[Host] In Bangladesh, Drinkwell has partnered with every utility company. But outside Bangladesh — in India, Cambodia, Africa — does the same model work?
[Guest] They're already doing it in West Bengal, India — another region hit hard by arsenic contamination. In 2025 they won the Assocham National Water Award, which suggests the model can replicate. But there's actually a deeper question underneath all this, and I think it's the one this episode really wants to leave with listeners —
[Host] What's that?
[Guest] If you live in a village on the outskirts of Dhaka, and a water ATM opens next to your house — the water is purified, the price is affordable — the question is: are you willing to walk ten minutes every day to fetch water, pay by phone, and let the system record your water-use data? That question is itself a kind of answer — the last mile of technology was never decided by the technology.
[Host] All right, that's it for today. You don't find fellow travelers in slogans — you find them in problems. This has been Local Action. See you next episode.