Published: 2023-09-15

I’ve been working on my ideas since high school, when I decided I would figure out a way to leave Siberia and go to the US. Some of the things I’ve built since then include a startup, a research program, and a 300-person family tree. The whole time I thought about how to make more positive impact on the world. That culminated in the last 6 months of my life, when I’ve done more building per unit of time than ever before. About 3 weeks ago, I realized I should probably start writing down the lessons I learn.

Think of this post as an incomplete, transitory, random-order brain dump of my learnings that I think will nonetheless be helpful to people who want to do good work.

This assumes you are excited about the problems you are working on.

  1. Solve a problem. Just make something people want and don’t make something people don’t want.
  2. There are always many problems you could work on. If you feel like you don’t see any problems, realize what constraints you are putting on your search that you can remove, then remove them and think again. If you are early on, pivot quickly if you realize you aren’t solving a good problem.
  3. If you are working on a good problem, stay with it a little longer than you’re comfortable. Often you’ll solve it by pushing just a little more.
  4. Things that increase energy: launching, learning a lot fast, people interacting with your work. Things that decrease energy: not launching, learning slowly or only a little, people not interacting with your work.
  5. If you do something hard, you’ll inspire people in ways you didn’t imagine.
  6. To get people to engage with your work, strive for one or both of these reactions:
  7. Invaluable things you can improve in no matter how good you are: learning how to motivate yourself, how to get out of tough places, how to go deep while going fast. Don’t overthink these. Just put in honest work to do them, then reflect, learn, and repeat.
  8. Very important to write and speak clearly. Clear writing and speaking comes from clear thinking and practice, so it’s important to think clearly and practice a lot. To practice writing, write a lot and publish. To practice speaking, record yourself speaking and force yourself to watch it.
  9. Recognize when first principles don’t give you the full answer and tribal knowledge is needed. Then learn from others as quickly as possible. Good examples are “how do I do X” and “what are the best tools to do Y”. See what people who worked on similar projects did. Don’t be afraid to ask, people are happy to teach and make recommendations. Go to the people you know, post on Twitter, LinkedIn, Reddit. Expect that most don’t know, but some will and all you need is to share what you’re working on, pick their brain, and be appreciative. Make sure the knowledge makes sense from first principles, then adapt it to your work.
  10. Asking others for help is a good way to sanity check yourself against thinking everyone knows X and you don’t.
  11. When you don’t understand / something doesn’t work and you need to solve it, just go deeper right away before you get too stuck and lose momentum. Understand the first principles, solve a similar but easier problem, and try again.
  12. If you’ve been thinking “I should try doing X, I wonder if that would be the right path for me” just go and do it. You don’t have to commit a lifetime, just go all in for 2 weeks. E.g. if you wonder whether you should do research, go do an independent research project. Feel like the kind of person who does research, think about research a lot, do something hard as part of your research, and see how you feel. You will learn more than you can imagine, solve your identity crisis, and naturally know what to do next.