Microsoft was lucky that it was not a capital intensive business. It was financed by money I had made in high school. I did the school scheduling. There were all sorts of things you could do with software to make money fairly easy in those days because the number of people who knew software was very small.

We knew software was this magical thing. Weirdly people didn’t understand that Moore’s law would mean computing power would be infinite. The best way to think about it was software would be the limiting factor towards any digitally assisted activity.

We had lots of customers who went bankrupt. I was hiring people who had children. They were moving to the city [for the job]. I always did this calculation: Do I have enough money that if no one pays me I can still pay the employees for a whole year?

[Why he recruited Steve Ballmer] I realized I needed somebody who could hire people and tell me not to sell things that weren’t done yet.

[Early Microsoft] Was a nice mix of young, naive, over optimistic people [who controlled everything] and the adults who could ask us to think twice before we did 3 crazy things per day. . .We [the young people at Microsoft] learned to accept the fact that they [the adults] weren’t quite as intense as we were. They had wives and kids. They would leave at some point during the day.

You can over worship and mythologize the idea of working extremely hard. It is true I didn’t believe in weekends. I didn’t believe in vacations. . .I don’t recommend it. I don’t think most people would enjoy it.

Once I got into my 30s I could hardly even imagine how I had done that. Now I take lots of vacations. My 20 year old self is so digested with my current self.

I have a fairly hardcore view that there should be a very large sacrifice made during those early years.

My greatest mistake ever is whatever mismanagement I engaged in that caused Microsoft to not be what Android is: The standard non Apple phone platform.

If you were in your 20s today what would you work on? AI. Problem number one in software is not solved. So that is what you would work on. . . I would be drawn to thinking about the structure of the software that can do the equivalent of what a human can do.

It is nice that the idea of being a founder is a thing. That you can meet other people doing it. It’s got a real ecosystem.

As a result you can move faster. You don’t need to invent certain things - like scaling with cloud computing. But you also get intense competition.

You have said that the most important issue is often not technical but cultural- it is convincing everyone that the company’s survival depends on moving as fast as possible. On the importance of a focus on velocity.

What is the most impactful unsolved problem that technology can solve that you’d like to see someone take on? This idea of creating an agent that helps a student like a one-on-one tutor would. There is nothing that matches a great one-on-one tutor teaching somebody. That is the gold standard that all other methodologies are compared to.

If there was a Microsoft of AI what would the business model be? A personal agent that sees all your information would replace all your applications. It is your portal to the world. It replaces social networking, search, productivity apps, going to amazon and shopping etc. So it has the opportunity to combine all those different elements - advertising, mark up etc. It would probably be done on a subscription basis for a lot of people so you don’t have to wonder about what the trade-offs are.

[Something I hope Bill is wrong about] The number of students who can learn without a social setting around it is very very small.

Learning is a weird social phenomenon.

Who is someone who has been instrumental to your success? Having people who have a little bit of distance is good. For the last 30 years I’ve gotten to be friends with Warren Buffett. He’s not in this tech world at all. Because he is not in this world he has a definite way of looking at things. Including this idea that work should be fun. He has made his work so much fun. He was a great counsel during the toughest time I went through [the antitrust lawsuit] That was a huge gift.

Full podcast here.

Learn from founders who came before you. Every week I read a biography of a founder and tell you what I learned on Founders podcast.

This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit davidsenra.substack.com