bratfarrar

AI can find bugs, and it can find targets for performance improvements. If you tell it to vibe code something that's fast, you're not going to get the results you want. Like the time Anthropic mostly but not quite created a compiler: https://www.anthropic.com/engineering/building-c-compiler. And if you want it fix things that are complex, you're going to have a bad time: https://1password.com/blog/why-ai-generated-patches-still-re...

I actively try to avoid discussing status. The 1:1 is primarily a meeting for them, whether or not they realize it. The goal in a 1:1 is primarily to get them talking about whatever it is that they want to talk about. Sometimes this just turns into chatting about non-work stuff. Sometimes we'll discuss problems they're having on the team. Sometimes they'll talk about work they want to do but haven't been given. Sometimes they'll have questions about "how the system works" (these are usually pretty interesting conversations). I try to tell stories from my own work history (frequently embarrassing ones) both because they're funny, and because it makes it easier for them to talk about their own potentially embarrassing problems. Sometimes I try to lead the conversation toward something that I want to talk about - like if I need to give them some feedback, or if I want to discuss their career development, or a new opportunity. Sometimes you can tell that they're circling around something they don't know how to bring up, and you can gently nudge them in the right direction.

Sometimes it's useful, and sometimes it isn't. Ultimately, if it helps to deepen trust and understanding, then that's a good result.

The implication is that humans have been trapped in a local maxima in terms of strategy and board reading. The question then becomes whether we can use this to bootstrap ourselves.

If it's free, you're the product.

Time for BST code golf - winner does it in the fewest characters!

Couldn't have said it better.

People will occasionally try that, but my point to them is that they should be writing a general purpose BST, which should also allow degenerate trees. So no, I do try to steer people away from theoretically correct but desperately inefficient approaches.

Hah! So awesome.

I wish I could upvote this about a thousand times.

:)

Agree completely.

True (I hope), true, and VERY true. :)

Ouch, not so much fun. I should have been much more specific that this post was about getting interviews at companies, not getting agencies to put you into their mill. There are some good recruiting agencies, but the problem is there's no way to know which is which unless you've worked with a lot of them and have context. You're right, this would be a good area for disruption (or maybe a good new feature for GlassDoor).

It's an interesting point - true "recruiters" are typically much more limited in their ability to evaluate resumes, which is why hiring managers do the resume review at TripAdvisor (I wasn't being precise in my usage). In reference to your point about universities mattering, though, I'd love to see more resumes from Penn... ;)

If you start at a large company with a reputation for only hiring top candidates, you will start your career with a higher salary, a great company on your resume, and experience in how world-class companies do things (both good and bad). These are your journeyman years, and will give you the experience, context, and contacts to be much more effective if and when you do decide to join a startup. It will also make you much more interesting and valuable to a startup (which will equal more equity).