You’re not going to find me advocating for letting the code go into production without review.

Still, that’s a different class of problem than the LLM hallucinating a fake API. That’s a largely outdated criticism of the tools we have today.

I’ve thought about this many times, and I’m just not seeing a path for juniors. Given this new perspective, I’m interested to hear if you can envision something different than I can. I’m honestly looking for alternate views here, I’ve got nothing.

I think it’ll just mean they they start their careers involved in higher level concerns. It’s not like this is the first time that’s happened. Programming (even just prior to the release of LLM agents) was completely different from programming 30 years ago. Programmers have been automating junior jobs away for decades and the industry has only grown. Because the fact of the matter is that cheaper software, at least so far, has just created more demand for it. Maybe it’ll be saturated one day. But I don’t think today’s that day.

Agents now can run compilation and testing on their own so the hallucination problem is largely irrelevant. An LLM that hallucinates an API quickly finds out that it fails to work and is forced to retrieve the real API and fix the errors. So it really doesn’t matter anymore. The code you wind up with will ultimately work.

The only real question you need to answer yourself is whether or not the tests it generates are appropriate. Then maybe spend some time refactoring for clarity and extensibility.

There are bad coders and then there are bad coders. I was a teaching assistant through grad school and in the industry I’ve interviewed the gamut of juniors.

There are tons of new grads who can’t code their way out of a paper bag. Then there’s a whole spectrum up to and including people who are as good at the mechanics of programming as most seniors.

The former is absolutely going to have a hard time. But if you’re beyond that you should have the skills necessary to critically evaluate an agent’s output. And any more time that they get to instead become involved in the higher level discussions going on around them is a win in my book.

But, when I do that, what I will actually be converting isn’t length to length. I’ll be figuring out how many sleepers per km, how many rail segments per km, how many buckets of spikes per km. None of those will be simple metric unit conversions.

This is actually the primary strength of imperial and the impetus behind most of its conversion ratios. Base 10 is just terrible for being divided. But if you have a mile of railroad, you can place your rail and stakes regularly at almost any foot-length and come out even.

My favorite use is actually just to help me name stuff. Give it a short description of what the thing does and get a list of decent names. Refine if they’re all missing something.

Also useful for finding things quickly in generated documentation, by attaching the documentation as context. And I use it when trying to remember some of the more obscure syntax stuff.

As for coding assistants, they can help quickly fill in boilerplate or maybe autocomplete a line or two. I don’t use it for generating whole functions or anything larger.

So I get some nice marginal benefits out of it. I definitely like it. It’s got a ways to go before it replaces the programming part of my job, though.

Sure you can move some parts of the conversation to a review session, though I think the answers will be heavily influenced by hindsight at that point. For example, hearing about dead end paths they considered can be very informative in a way that I think candidates assume is negative. Nobody expects you to get it right the first time and telling the interviewer about your binary tree solution (that actually doesn’t work) can be a good thing.

But the biggest problem I think with not being in the room as an interviewer is that you lose the opportunity to hint and direct the candidate away from unproductive solutions or use of time. There are people who won’t ask questions about things that are ambiguous or they’ll misinterpret the program and that shouldn’t be a deal breaker.

Usually it only takes a very subtle nudge to get things back on track, otherwise you wind up getting a solution that’s not at all what you’re looking for (and more importantly, doesn’t demonstrate the knowledge you’re looking for). Or maybe you wind up with barely a solution because the candidate spent most of their time spinning their wheels. A good portion of the questions I ask during an interview serve this purpose of keeping the focus of the candidate on the right things.

I’m not sure that offline or alone coding tests are any better. A good coding interview should be about a lot more than just seeing if they produce well structured and optimal code. It’s about seeing what kinds of questions they’ll ask, what kind of alternatives and trade offs they’ll consider, probing some of the decisions they make. All the stuff that goes into being a good SWE, which you can demonstrate even if you’re having trouble coming up with the optimal solution to this particular problem.

First thing you do? Unzip.

I think when you consider the rate of advancement of any technological species, “roughly the same level as us” basically implies that they got started at exactly the same time. Even an extra thousand years of technological advancement would put them far ahead of us. A million years would put them unimaginably far ahead.

On a cosmic scale, that’s nothing. That’s a tight window and given the like 8 billion years that planets with the required elements have had to form, I would doubt that no other species had a chance to surpass us.

I mean, Agile doesn’t really demand that you do or don’t use tickets. You can definitely use tickets without scrum.

deleted by creator

That’s… the point? Civilizations with that kind of tendency may very well destroy their planet and/or themselves long before they advance to the point where they are detectable to an outside observer many light years away.

The human race is at the moment in a race against time. We’re hoping that we can develop new technology to save ourselves faster than we destroy everything around us. This kind of race has probably happened countless times across the vast universe and perhaps the laws of physics ultimately make the race unwinnable. These laws limit how much technology can do for any species, no matter how smart, so it would be a universal filter.

If the only way to win the race is to slow down the destruction of the environment to the point that the species is undetectable, that solves the Fermi paradox.

Those… don’t hold any weight lol. Once you post on any website, you hand copyright over to the website owner. That’s what gives them permission to relay your message to anyone reading the website. Copyright doesn’t do anything to restrict readers of the content (I.e. model trainers). Only publishers.

I’m guessing that 6-20 minutes is like the actual time spent driving screws or drilling holes. Each one takes maybe a few seconds. 6-20 minutes in that case translates to hundreds of screws driven, even on the low end. So not nearly as worthless as the time makes it sound.

If we’re in a simulation, it’s probably a massive universe-spanning one. We’re just a blip, both within the scale of the space of the universe and within the history of time of the universe. In that case, we’re not important enough for a simulation creator to even care to adjust our capabilities at all. They’re not watching us. We’re not the point of the simulation.

Everyone’s talking about encyclopedias but they weren’t always that useful either. They can only fit so much information in those books so some topics would only get like 3 sentences dedicated to them. So yeah, if you were writing a research paper for school you’d spend lots of time at the library trying to find books that had another smidge of information you needed.

If you were lucky, you’d find a really good book that was very relevant to your topic and lean heavily on that. Otherwise, you’d wind up with like a few sentences each from a dozen books that you have to tie together somehow. Wasn’t fun.

It’s not the free fall that I’m worried about. It’s what comes after.

I’ll tell you why I say the answer is no to whether or not the occult, demons, or ghosts exist. There’s a phenomenon in statistics where if you were to select a random element from an infinite set of equally probable elements, the probability that a specific element will be selected is 0%. Not close to 0, literally 0.

These kinds of supernatural phenomena that have no evidence belong to an infinite set of equally unlikely phenomena with no evidence. Their likelihood of being real is 0%. Only when phenomena has some tangible evidence explaining it can we elevate it to a finite set with a non-zero likelihood of being real.

Sorry but I’m going to call out what I see as some pretty blatant motte-and-bailey argumentation by the OP and their offense taken to people trying to nail down the definition of supernatural is illustrative.

They have their bailey, belief in things like the occult, ghosts, demons, etc, that are almost certainly bullshit. To the extent that they can be falsified, they have been. This is the typical definition of what people think when you say “supernatural” and people are right to answer “no” when asked if they believe in it.

But then you have OP falling back on their motte when this happens, taking a nebulous definition of supernatural and asking rhetorical philosophical questions about reality, perception, and the unknown. The fallacy is that these questions do nothing to strengthen or refute the original argument about the supernatural.

Nobody is here to argue that nothing is unknown and even unknowable but that doesn’t make the things that people typically call “supernatural” any less bullshit. Demons and ghosts are just not the kinds of things that are waiting around to surprise us. And shifting the conversation from your bailey to your motte to protect your feelings on the former is not a good way to have a friendly debate.

All that aside, if you are interested in expanding your understanding of the universe then I’d really encourage you to divert the effort you’re putting into the “supernatural” into learning about the actual natural universe instead. Our universe really is fantastic on its own. There’s plenty of interesting, wacky, and unknown things happening all around us that you can learn about without resorting to magic. If anything, magic is the boring answer imo.