About
Every era has a culture, and every culture is a marriage.
The last century married science to industry. This one is marrying art, science and code, and for the first time the machine is a participant in all three. I have worked in all three for twenty-eight years, usually at the same time. Marain is where that gets applied to a business, and then built.
Three threads, twenty-eight years. Scroll, and travel along them.
- Building · 1998
Pascal at thirteen, then a company.
My first program was in Pascal, at thirteen. I wrote C and Clipper for pocket money through school, and in 1998 I turned that into a company that built more than a hundred websites on a content management system I wrote by hand. That system was acquired. The dot-com crash took the rest.
- Writing · 2001
A newsroom, from conflict zones to dog shows.
I left my first company for a newsroom in 2001 and spent the next six years reporting on the human condition, from conflict zones to dog shows. At Hindustan Times I launched the first science-and-technology section in a mainstream Indian newspaper, and covered the Indian Ocean tsunami, the Nepal coup, and the opening of the first road between the two halves of Kashmir.
- Writing
A science show, a national desk, a channel built from nothing.
At CNN-IBN I produced and anchored my own weekly science show, Connect, and worked the national desk covering politics and the political economy alongside science and technology. At NewsX I was one of the youngest executive producers in Indian television, part of the small team that planned and built the channel, hired into it, and put it on air.
- Writing
Three newsrooms joined into one.
At Mint I built the video desk and helped join the print, online and video newsrooms into one. Its founding editor, Raju Narisetti, calls that the world’s first print, online and video integrated newsroom.
None of it was a detour. The habit a newsroom trains is deciding what a complicated thing actually means and then saying it plainly enough that someone can act on it before the deadline, which is most of what a positioning problem is.
- Measuring · 2008
Numbers and people, read at the same time.
In 2008 I co-founded an analytics firm doing political and market research, quantitative and qualitative together, for politicians, property developers, consumer goods companies and funding agencies. The idea that you read the numbers and the people at the same time is not something I picked up when machine learning got fashionable.
- Measuring
Instruments, not reports.
Through the agency years I built instruments rather than reports: search-path analysis that reconstructed what a buyer was actually trying to find out, review mining across the whole population of reviews instead of a sample, and a map of millions of articles built to show where nobody was talking.
- Measuring
A metric in every cell.
For the Gates Foundation in India I wrote a measurement framework with a named metric in every cell of it, and stated in the document that no benchmarking data existed yet, so the framework’s first job was to generate its own. That is the whole method in one page.
- Building · 2018
Where a gate decides.
When the Statue of Unity opened in 2018, its ticketing and gates ran on software I wrote in two days: C on a Raspberry Pi at every gate, each reporting to a central hub that recorded every transaction. It sat where a person arrives to look at a work of art and code decides whether the gate opens.
- Measuring · 2019
Twenty-nine countries, nothing sent to a server.
In 2019 my company was analyzing narrative across twenty-nine European countries, reducing concept space into maps a human analyst could read, with all the processing happening in the browser and no data transported to a server. Data sovereignty was an engineering decision I was willing to pay for seven years before the market started asking about it.
- Measuring
The wrong call.
Told in full below, as the log entry it became.
- Building · December 2023
A streaming platform, acquired.
Later I co-founded a streaming platform that let media companies launch a full service in weeks rather than years, and it was acquired in December 2023.
- Building · 2026
Marain runs on the same habit.
Marain itself runs on the same habit. 34 automations at a 0.47% failure rate, private models on hardware I own, a semantic search index across two domains, and an ops console I use every day. None of it is for sale. It is there so you can check that the advice runs.
The record
Four companies founded, two sold.
- 4
- companies founded
- 2
- sold: Laminar, Elephants & Monkeys
- 2009
- coaching early-stage founders since
- 10→100
- people in the India digital practice at Publicis, one office to four
- 2,000
- people across six Asian markets, in the integrated-planning remit I led
- $20M
- a year of P&L, run directly
Laminar, the streaming platform business, was acquired by APMC in December 2023. Elephants & Monkeys, the digital agency I founded in 2008, was acquired into 2020 Social and then into MSLGROUP, part of Publicis Groupe, in 2010. My first company, in 1998, is the origin story rather than an exit: I wrote a CMS that was acquired, and the dot-com crash took the rest.
I have been coaching early-stage founders since 2009.
At Publicis I took the India digital practice from ten people to more than a hundred and from one office to four, led a 200-person India team, and then led integrated planning across every Publicis agency in the region, an organization of close to 2,000 people across six Asian markets. I have carried a P&L of roughly twenty million dollars a year, run directly. The clients were some of the largest brands in the world: P&G, Coca-Cola and Sony Mobile among them, across beauty, beverages, consumer electronics and beer in a dozen markets.
At Xynteo I was Chief Digital Officer, advising Unilever and Shell among others on how they went to market and how they talked about it.
Before all of that: six years in newsrooms, print and television.
The name
Marain is a designed language.
The paragraph on the right arrives in one and resolves into the other. That is the job.
That is what this practice does between art, science and code, and between the people who fund a thing and the people who build it.
The wrong call
I overrode the model.
That is why every system I build now logs a human override with a written reason.
I’m wrong in public when I’m wrong. That is the whole point.
Narendra Nag. info@marain.space