How a year of lessons fits into one day of building

It looks like being a founder turns you into a salesperson in unexpected places.

I happened to be in SF today for business, and an informal meeting with Spandan Madan and his team, who are working on magically making it easier to manage our memories, was a chance to make the most of my time there. I feel a bit guilty: the idea was to learn how their journey is going, and I ended up demoing my product.

A couple of interesting questions came up. How did I build so much so fast? And why build everything in one go?

Why everything in one go

I already know the answer to the second one. Chasing completeness or perfection is a classic temptation for engineers. It's probably better to go with the flow.

How so much so fast

The first question is a lot more important. A feature I implement in one day is the result of something I built or saw in the past that took three to six months of planning, prioritization and user research. Then came another three to six months of design, with the added constraint that you often can't choose the best long-term design, because the organization's structure has locked you into certain choices.

To be fair, changing an application that's already serving users in production makes you more careful and considerate. All of that is a cost.

Experience lets you squeeze a year's worth of that journey into a better feature in a single day.

Could I repeat it in another domain?

Probably not to the same extent. I'd go through multiple iterations before arriving at a good solution. That's a tax that has to be paid.