Forty years in, the interesting question is no longer what has been built. It is what should be built next, and by whom.
I have spent a career answering the question of what to build. I now spend most of it on the question of what to leave behind.
The next horizon is not mine to cross. It belongs to people who are currently in classrooms, or three years into a first job, or running something small that nobody has noticed yet. My work now is to make their starting position better than mine was — with capital where it is useful, advice where it is asked for, and institutions that will still be standing when they arrive.
Five things will decide whether that generation inherits momentum or repair work. They are set out below. None of them is complicated. All of them are slow.
Rakesh Sharma
Building for people I will never meet.
The decisions that matter most are the ones whose consequences land after you are gone. What we choose to construct, deplete, or protect now becomes the starting position for people who had no say in it.
From a department to a constraint.
The next decade will separate organisations that reported on sustainability from those that let it change what they were willing to build. Only the second group will still be defensible in 2040.
New tools, older questions.
AI and digital systems have changed the cost of trying things, not the standard for whether they should exist. The advantage now goes to those who can hold both: speed of execution and slowness of judgement.
Infrastructure is a promise.
Every road, grid, and institution is a commitment made to a population about what will still be working in thirty years. India's next chapter depends less on ambition than on maintenance — on finishing what was started.
The only compounding asset.
Capital is replaceable and technology is copyable. Capability inside people is neither. Every serious plan for the future eventually reduces to whether enough people were given the chance to become good at something.
If you are building something that belongs to one of these five, there is likely a conversation worth having.