Thinking, written down slowly.

Mrineh Group began with a simple conviction: a business should improve the life it touches and remain useful long after its first success. I do not see our companies as a collection of unrelated ventures. I see them as different expressions of the same responsibility—to build carefully, serve honestly, and think beyond the immediate quarter. The Yamuna Resort was the beginning in 2016. Its pause during the pandemic taught me that endurance is not the same as uninterrupted growth. Sometimes endurance means protecting the original purpose until the right moment to continue. Mrineh followed in 2021 because education seemed to me one of the clearest places where patient work could matter. Students need academic knowledge, but they also need judgement, communication, emotional steadiness, financial understanding, and the confidence to recover from failure. Ardevaé extended the same philosophy into furniture: respect the material, respect the craft, and make objects intended to remain. ODCorrect applies it to clothing designed for real life rather than a passing season. Mrineh AI brings it into technology, where intelligence should reduce noise, return time, and behave with restraint. These companies work in different fields, but the standard is shared. We want to create things people can trust because the substance is stronger than the announcement. That means accepting slower decisions when quality requires them, documenting how work is done, and building teams that do not depend on one person carrying every answer. A founder can begin an organisation, but only principles, systems, and capable people can turn it into an institution. My ambition for Mrineh Group is therefore not simply to own more companies. It is to build a house in which each company can develop its own character while remaining accountable to a common set of values: curiosity, patience, craft, responsibility, consistency, execution, kindness, and long-term thinking. Growth matters, but not at the cost of usefulness. Technology matters, but not when it takes more attention than it returns. Brand matters, but only when it is supported by the quality of the work. The measure I care about is whether these organisations continue keeping their promises—to students, guests, customers, craftspeople, colleagues, and communities—over many years. That is the work of building Mrineh Group. It is quieter than a launch and slower than a campaign. It is the daily practice of making decisions we can still defend a decade from now, so the institutions eventually become larger than the person who started them.

Speed is the easiest thing in business to buy and the easiest thing to mistake for progress. You can buy a launch date, a growth curve, an office full of people and none of it tells you whether the thing underneath is actually built. I have watched companies move very fast in the wrong direction and call it momentum. What I trust instead is compounding: a product that is slightly better this quarter than last, a team that has been together long enough to disagree well, a reputation assembled one kept promise at a time. None of that photographs well. All of it is nearly impossible to compete with after a decade. Borrowed speed always comes with an invoice shortcuts in materials, in hiring, in the parts of the work nobody sees and the invoice arrives late, usually when you can least afford it. So some businesses should deliberately move slower. A school programme, a piece of furniture, a place people sleep: these are promises about the future, and a promise made in a hurry is usually a promise made carelessly. Building slowly is not the absence of ambition. It is ambition with a longer memory.

A score describes one afternoon. It says nothing about whether a student can hold a difficult conversation, sit with a problem that does not resolve in forty minutes, notice when they are wrong, or manage the first salary they ever earn. We have built an entire system around measuring the narrow part and hoping the rest arrives on its own. It rarely does. The skills that decide most lives judgement, communication, emotional steadiness, money sense, curiosity that survives being bored are teachable, but only if someone decides to teach them. Failure especially: a young person who has never been allowed to fail in a low-stakes room will be introduced to it later, in a room where it costs a great deal more. I want education to produce capable human beings rather than well-decorated report cards. That is the whole argument behind the work we do in schools. Not a replacement for academics a completion of them, so that the syllabus finally includes the part of life the examination never prints.

Most of the decisions I am proudest of looked mildly irrational when I made them. Choosing the slower supplier. Turning down the growth that would have required us to lie a little. Paying for the material nobody would ever see. On a ten-week horizon these are mistakes; on a ten-year horizon they are the only reason anything still stands. Durability is unfashionable because it produces no news. But hype depreciates and reputation appreciates, and the gap between them widens every year you stay honest. There is a particular kind of product that becomes more valuable with time a chair that has been sat in for fifteen years, a school programme whose first students are now adults you can point to, a place that people return to with their children. You cannot manufacture that. You can only survive long enough to have it. Delayed gratification is not patience for its own sake; it is a bet that the market eventually notices who was telling the truth.

Most software is designed to be needed rather than useful. It notifies, nudges, celebrates, re-engages it behaves like something that is afraid of being forgotten. I think the better standard is the one we hold a good colleague to: attentive when required, silent otherwise, and never demanding credit for doing its job. Quiet technology reduces the number of decisions a person has to carry. It removes friction instead of relocating it. It closes cleanly. Applied to artificial intelligence, this matters more, not less: a system that can do anything is under enormous temptation to do everything, and the result is a companion that competes for attention rather than protecting it. The measure I care about is simple and slightly unfashionable did the person get time back, and did they leave with their attention intact? Software that answers yes to both is doing something genuinely intelligent. Software that is merely loud is doing something else.

A startup is an experiment; an institution is a promise. The difference is not size or age, it is whether the thing can continue without the person who began it. That requires unglamorous work: writing down how decisions are actually made, building systems that do not rely on one person remembering, hiring people who are allowed to be better than you at their part, and treating culture as something maintained rather than announced. Succession is the honest test. If the founder leaving would break the company, then the company is really a very elaborate description of one person's schedule. I take that seriously because of what our businesses touch a student's year, a family's home, someone's holiday. Those obligations should not expire with my attention span. Legacy is a heavy word and I use it carefully; what I mean by it is ordinary and specific. Build the thing well enough, and staff it honestly enough, that it is still keeping its promises long after nobody remembers who made them.