Cursor Team: Future of Programming with AI | Lex Fridman Podcast #447 (transcript)

Michael Truell · 6 Oct 2024

Michael Truell did not write this page.

Every claim below was made in this piece, quoted word for word and numbered in the order the piece makes them, so you can read it there rather than take our word for it. The sentence above each quote is our reading of the claim, not their wording. Each quote was checked against a stored copy of the page at build time; where the two differ, the quote is the fact.

Our reading
If AI builds more and more of a system, it needs to verify as well as generate, including checking its own code.

Their wordsIf AI is building more and more of the system for you, you need to not just generate but also verify. And without that, some of the problems that we’ve talked about before with programming, with these models will just become untenable. So it’s not just for humans like you write a bug, I write a bug, find the bug for me, but it’s also being able to verify the AI’s code and check it is really important.
Our reading
Automatic context given to coding models should clear a high bar for accuracy and relevance, since more context makes them slower, costlier and more confused.

Their wordsSo the bar for accuracy and for relevance of the context you include should be quite high.
Our reading
The ceiling for AI coding products is very high: the best product in three to four years will be much more useful than the best one today.

Their wordsI think that the ceiling here is really, really, really incredibly high. So I think that the best product in three to four years will just be soon much more useful than the best product today.
Our reading
In AI coding products, brand and moats matter less than continued innovation: stop innovating on the product and you lose.

Their wordsI think in the end, just if you stop innovating on the product, you will lose.
Our reading
The programmer should stay in the driver's seat for a long time, with speed, agency and control.

Their wordsI think we’re really excited about a future where the programmer is in the driver’s seat for a long time. And you’ve heard us talk about this a little bit, but one that emphasizes speed and agency for the programmer and control.
Our reading
The best engineering involves many tiny decisions about what exactly to build and hard trade-offs between speed and cost.

Their wordsBut I think a lot of the best engineering, the engineering we enjoy, involves tons of tiny micro decisions about what exactly you’re building, and about really hard trade-offs between speed and cost and just all the other things involved in a system.
Our reading
This is a really exciting time to be building software, and programming today is more fun than it was in 2012 or 2013.

Their wordsI actually think this is a really, really exciting time to be building software. We remember what programming was like in 2013, 2012, whatever it was. And there was just so much more cruft and boilerplate and looking up something really gnarly. And that stuff still exists. It’s definitely not at zero. But programming today is way more fun than back then.
Our reading
As AI changes programming, people's taste and creative ideas will be magnified and boilerplate text editing will matter less.

Their wordsI think that people’s taste and creative ideas will be magnified. And it will be maybe less, a little bit, about boilerplate text editing.

Keep reading

The other pieces are the ones whose claims are worded most like this one's. That measures language, not agreement: they may argue the opposite.