korrents

korrents · video

Claude Code creator Boris Cherny on the end of the software engineer

Boris Cherny · 1h 02m · youtube.com ↗

30 korrents from this recording

1h
Boris Cherny did not write this page.

Every claim below is a statement made in this recording, quoted word for word and linked to the second it was said, so you can hear it rather than take our word for it. The wording comes from the transcript published alongside the recording; the sentence above each quote is our reading of the claim, not their wording.

  1. 0:21:35 · watch on youtube.com↗

    exponentials are very hard to think in. So honestly, everyone that's saying that they know, no one actually knows. We're all guessing.
  2. 0:21:46 · watch on youtube.com↗

    One is there's going to be a lot of companies that need less engineers because engineers are more productive. So you just don't need as many engineers to do the same work.
  3. 0:21:54 · watch on youtube.com↗

    I think at the same time, there's going to be a lot of companies that need a lot more engineers because every engineer is more productive. The company could do more things.
  4. 1 min later
  5. 0:22:48 · watch on youtube.com↗

    And so this is what makes me think that over time, if you kind of project this trend a little bit, what's going to happen is everyone that's not an engineer is going to code a little bit more.
  6. 0:23:07 · watch on youtube.com↗

    Like we can call it a builder, we can keep calling it an engineer, we can call it a product manager. I don't know what the title is, but the role is changing.
  7. 2 min later
  8. 0:24:46 · watch on youtube.com↗

    And I think the thing that we're seeing right now is like this on a speedrun. And, um, you know, it's, but it's sort of the same thing. We're, we're hitting like very similar issues.
  9. 4 min later
  10. 0:28:19 · watch on youtube.com↗

    And I think this is what we're starting to see with a lot of companies that are really getting it, is they just put Claude exactly at the center of it. It's not like a thing on the outskirts somewhere.
  11. 2 min later
  12. 0:30:32 · watch on youtube.com↗

    And when you think about the kinds of stuff that engineers do, coding is a small percent of it.
  13. 1 min later
  14. 0:31:13 · watch on youtube.com↗

    when the model does the coding, they're freed up to do all this other stuff that they actually enjoy doing a lot more, like, you know, like talking to users and like figuring out what's next.
  15. 1 min later
  16. 0:32:23 · watch on youtube.com↗

    You know, like our team is an early indicator of what's happening in engineering. Engineering is an early indicator of what happens to everything outside of engineering.
  17. 1 min later
  18. 0:33:19 · watch on youtube.com↗

    I think for me, this is part of like a much broader transition and it's not about atrophy at all. It's just about like programming is always a thing that is in flux.
  19. 1 min later
  20. 0:34:11 · watch on youtube.com↗

    And now it's like changing again. It's now you talk to the agent and it's actually, I think, about to change one more time where you talk to an agent that talks to agents that does the coding.
  21. 2 min later
  22. 0:35:46 · watch on youtube.com↗

    But I think in almost every other case, it's sort of like, maybe it's like a honeymoon period where you kind of get used to the model and, you know, at first it's magical and then you kind of get used to it.
  23. 0:36:00 · watch on youtube.com↗

    But I don't think it's really about craft because the model's code at this point is just much better than the code I would have written.
  24. 4 min later
  25. 0:39:33 · watch on youtube.com↗

    And because the time was freed up, now you have this choice of what you want to do with this time. And I think it's kind of similar right now. It's like, it's similar for any technology. It gives you more choice.
  26. 1 min later
  27. 0:40:24 · watch on youtube.com↗

    but actually, if you're at all entrepreneurial, go start a startup. It's just like there has never been a better time in history to start a startup.
  28. 1 min later
  29. 0:41:24 · watch on youtube.com↗

    I mean, there's like a version of it where like the idea of jobs kind of doesn't make sense anymore. The idea of companies doesn't make sense. The idea of software doesn't make sense.
  30. 1 min later
  31. 0:42:05 · watch on youtube.com↗

    I don't think we're going to call them engineers, but if we talk about like people writing code or people like using agents, you know, like people using Claude to write, to write the code, I think there will be 100 times more engineers than there are today.
  32. 3 min later
  33. 0:45:00 · watch on youtube.com↗

    And every night I have like, you know, hundreds or sometimes thousands of agents that are just running 5, 10, 20 hours. And this is just how engineering is done now.
  34. 1 min later
  35. 0:45:57 · watch on youtube.com↗

    But I think once you actually use the technology and like, like you, like, use it every day, you kind of feel like what's, you know, what's missing because you feel it.
  36. 4 min later
  37. 0:50:13 · watch on youtube.com↗

    I think that, you know, broadly, this is not a problem that we can solve. This is bigger, you know, than one company. you actually really don't want one company to solve it because it could be the wrong solution.
  38. 1 min later
  39. 0:51:27 · watch on youtube.com↗

    one of the biggest arguments was we want people to experience it so that they can understand it, so that they can kind of play a part in figuring out like as a society what we should do.
  40. 0:51:37 · watch on youtube.com↗

    But if you like keep this technology locked away, so, you know, no one knows what it can do and no one can experience it, it's much harder for people to form a point of view about it.
  41. 2 min later
  42. 0:53:48 · watch on youtube.com↗

    One of the ways that I've seen work the best is you just give everyone tokens. You make everyone feel safe experimenting, and the ideas will come from the people that you don't expect a lot of the time.
  43. 1 min later
  44. 0:54:33 · watch on youtube.com↗

    it's not necessarily the people that are, you know, the most productive and the best with the tools of today that are going to be the best with the tools of tomorrow.
  45. 1 min later
  46. 0:56:00 · watch on youtube.com↗

    If you have an app that gets more valuable with more people being on the app, it doesn't matter who's building the app or whether AI exists or is an important force or not. That's still just as important.
  47. 0:56:22 · watch on youtube.com↗

    So like, you know, you're on vendor A and you want to move to vendor B, actually Quad can probably just move you from A to B. And like the switching cost is not a giant moat anymore.
  48. 1 min later
  49. 0:57:07 · watch on youtube.com↗

    I think there's going to be a lot of new ideas coming from not big companies, but tiny startups of like 1 or 2 or 10 people. And I think the number of these startups is going to explode.
  50. 1 min later
  51. 0:58:17 · watch on youtube.com↗

    It's like someone that knows the domain really well and has an army of clods can just do so much more than, you know, what they could have done before, even if they had like an army of humans doing it.
  52. 2 min later
  53. 1:00:42 · watch on youtube.com↗

    And the only way to make it better is to listen to people, especially when they say something doesn't work and to like keep improving it and keep making it better. And this is actually the way that awesome product is built.