DERIN GEORGE — FOUNDER
Why I’m building
Thalwag.
Not a bio. Not a CV. A reason.
I grew up on the coast and couldn’t stop thinking about the water.
I’m from Kerala. I grew up near the sea. The Arabian Sea was not an abstraction to me; it was the thing you looked at in the morning, the thing that set the weather for the day, the thing the fishers read before they decided whether to go out. There are communities on this coast who have been doing this for centuries. They know the sea in a way that no instrument has ever fully captured — the colour of the water before a squall, the way certain swells run before a storm, the aggregation patterns that tell you where fish will be three days from now.
What I couldn’t stop thinking about, once I started paying attention to the oceanographic literature, was the gap. The Indian Ocean — one of the most climatically important bodies of water on earth, the engine of the Asian monsoon, the fishing ground for hundreds of millions of people — is among the least-observed major ocean basins in the world. The data that scientists and forecasters and fisheries managers need to do their work is not there. Not in adequate quantity. Not at adequate resolution. Not continuously.
And then I looked at what was already at sea: 200,000 fishing vessels, operated by people with extraordinary environmental knowledge, crossing the most data-sparse waters in the Indian Ocean every single day.
I couldn’t believe no one had built this yet.
That is not a credentials statement. I don’t have a career in oceanography. I don’t have a navy or a government mandate or a consortium of institutions behind me. What I have is clarity about the problem, enough technical knowledge to see that it is solvable, and an honest inability to walk away from it. That is probably not sufficient. But it is what I have, and I would rather start with insufficient resources than not start.
The name Thalwag comes from the deepest channel of a river — the line of fastest flow. I chose it because it feels accurate: I am trying to find the deepest line through a complicated problem, the path that is most direct, most efficient, least likely to get stuck in eddies. That is the aspiration. Whether I achieve it is a different question.
One person can now start what once needed a navy.
Three things changed in the last decade that make this possible in a way it wasn’t before, and I want to be honest about what they are, because the timing is not incidental.
Sensors became cheap enough.
Solid-state ocean sensors that once cost thousands of dollars per unit — and required specialist handling — now cost tens of dollars and can be operated by anyone with basic training. The price collapse is not complete and it is not uniform across all sensor types, but for temperature, salinity, and dissolved oxygen the economics shifted decisively in the last five to eight years. A network that would have cost tens of millions of dollars to equip a decade ago costs a fraction of that now.
The models became open.
Global ocean analysis systems that previously existed only inside national meteorological agencies — and were accessible only to researchers with institutional agreements — are now publicly available. The Copernicus Marine Service provides daily global ocean analyses free of charge. NEMO, the numerical ocean model, is open source. The data assimilation frameworks that turn raw observations into coherent ocean state estimates are documented and, in some cases, openly available. One person with the right technical background can now access the modelling infrastructure that once required a large government programme.
The urgency became undeniable.
The Indian Ocean is warming faster than any other major ocean basin. The patterns of monsoon variability that South Asian agriculture has organised itself around for centuries are shifting in ways that are not yet fully understood, in part because the observational record is too sparse to characterise them. The fisheries that hundreds of millions of people depend on are responding to thermal and oxygen conditions that we are measuring inadequately. This is not a future problem. It is happening now, and we are trying to manage it with insufficient data.
None of these three changes, on their own, is sufficient. Cheap sensors without a model are just noise. An open model without observations to feed it produces an uncertain estimate of an uncertain ocean. Urgency without a concrete plan produces good intentions and nothing else. But the combination — cheap sensing, open modelling, genuine urgency — creates an opening that did not exist in the same form before. I intend to use it.
THE MISSION
Turn India’s fishing fleet into the world’s largest ocean-observation network and a living, self-correcting model of the Indian Ocean — operated in partnership with the fishers who make it possible, and open to everyone who needs it.
That’s it. Everything on this site — the science, the open data, the partnership model, the commercial products — is in service of that sentence. If I’m still working on this in twenty years, it will be because I think it matters enough to spend a working life on. I do.
WRITE TO ME DIRECTLY
I read everything. I respond to things that engage with what Thalwag is actually building. If you’re a researcher, an oceanographer, someone from a fishing community, a government official, an investor, or someone who just finds the problem interesting — write. I’d like to hear from you.
derin@thalwag.comKochi, Kerala, India ·9.9312° N, 76.2673° E
BUILT IN THE OPEN
Thalwag is a work in progress. The science is documented in a public preprint. The observation data are published under CC BY 4.0 as they are collected. The methods are reproducible. The code is public. The arguments made on this website are the same arguments made in the scientific paper: specific, testable, and honest about where I don’t yet know the answer.
I’m building this in public because I think that’s how it should be built. If you find a mistake — in the science, in the argument, in the data — I want to know.
— Derin
Kochi, 2026