I live in a fairly expensive city in the UK. Working minimum wage for 2 weeks will pay for a room in a flat share, plus my households food and required bills.

It’s not much of a life but the same still stands in many cities.

To back up the other commenter - it's not the same. https://godbolt.org/z/r6e443x1c shows that if you write imperfect C++ clang is perfectly capable of optimizing it.

What's strange is I'm finding that gcc really struggles to correctly optimize this.

This was my function

    for (auto v : array) {
        if (v != 0)
            return false;
    }
    return true;

clang emits basically the same thing yours does. But gcc ends up just really struggling to vectorize for large numbers of array.

Here's gcc for 42 elements:

https://godbolt.org/z/sjz7xd8Gs

and here's clang for 42 elements:

https://godbolt.org/z/frvbhrnEK

Very bizarre. Clang pretty readily sees that it can use SIMD instructions and really optimizes this while GCC really struggles to want to use it. I've even seen strange output where GCC will emit SIMD instructions for the first loop and then falls back on regular x86 compares for the rest.

Edit: Actually, it looks like for large enough array sizes, it flips. At 256 elements, gcc ends up emitting simd instructions while clang does pure x86. So strange.

I;ve had to coerce gcc to emitting SIMD code by using int instead of bool. Also, the early return may be putting it off.

Doing both of those things does seem to help: https://godbolt.org/z/1vv7cK4bE

GCC trunk seems to like using `bool` so we may eventually be able to retire the hack of using `int`.

I see yeah that makes sense. I wanted to highlight that "magic" will, on average, give the optimizer a harder time. Explicit offset loops like that are generally avoided in many C++ styles in favor of iterators.

Even at a higher level of abstraction, the compiler seems to pull through: https://godbolt.org/z/1nvE34YTe

It emits a cmp/jmp still when arithmetic would be fine though which is the difference highlighted in the article and examples in this thread. It's nice that it simplifies down to assembly, but the assembly is somewhat questionable (especially that xor eax eax branch target on the other side).

Hot reloading for development loops is _the_ canonical use case for this.

If people are realistically a baseline of 10x more productive where are all the features, games, applications, SAAS’s that are suddenly possible that weren’t before?

They're out there. I've noticed a surge of Show HNs right here. A lot of it is vibe coded.

I would like to see GitHub's project creation and activity charts from today compared to 5 years ago. Similar trends must be happening behind closed doors as well. Techy managers are happy to be building again. Fresh grads are excited about how productive they can be. Scammers are deploying in record time. Business is boomin'.

It's likely that all this code will do more harm than good to the rest of us, but it's definitely out there.

Windows 11

AI might be 100x faster than me at writing code - but writing code is a tiny portion of the work I do. I need to understand the problem before I can write code. I need to understand my whole system. I need to test that my code really works - this is more than 50% of the work I do, and automated tests are not enough - too often the automated test doesn't model the real world in some unexpected way. I review code that others write. I answer technical questions for others. There is a lot of release work, mandatory training, and other overhead as well.

Writing code is what I expect a junior or mid level engineer to spend 20% of their time doing. By the time you reach senior engineer it should be less (though when you write code you are faster and so might write more code despite spending less time on it).

My experience is very different. A junior should be spending >90% of their time coding, a mid level about 75% and a senior about the same. It only really splits after that point. But the senior spending 25% more time on the wrong thing is worse than than them spending 25% less time on the right thing.

It’s only when you get to “lead” (as in people manager) or staff (as in tech unblocked) you should spend the rest of your time improving the productivity of someone who spends more of their time coding than you do.

I can tell you, with absolute certainty, that before AI ~0 junior/mid level devs spent just 20% of their time programming. At least not at tech companies

Every company I've worked for that is the case. Note that I divided a lot of the work out that might be counted as coding though.

For a very small number of people the hard part is writing the code. For most of us, it’s writing the correct code. AI generates lots of code but for 90% of my career writing more code hasn’t helped.

> you do end up taking on more than you can handle, just because your mob of agents can do the minutia of the tasks, doesn’t free you from comprehending, evaluating and managing the work

I’m currently in an EM role and this is my life but with programmers instead of AI agents.

Also EM and it feels like now I have a team of juniors on my personal projects, except they need constant micromanaging in a way I never would for real people.

Does AI write 100% correct code? No, but under my watch it writes code that is more correct than anything that anyone else on the team contributed in past year or more. Even better when it is wrong I don’t have to spend literal hours arguing with it nor I have to be mindful how what I’m saying affects others feelings so I get to spend more time on actual work. All in all it’s a net positive

I agree.

I provide specific instructions, gotchas when prompting the Agent to write the code. I churn out more instructions quickly by using my voice.

Yes it makes mistakes, but it can correct them quickly as well. This correction loop takes more time if it is a human in my team working.

I never said it’s not a net positive - I said that writing more code won’t solve the problem.

> under my watch it writes code that is more correct than anything that anyone else on the team contributed in past year or more

This I don’t believe.

> For most of us, it’s writing the correct code.

I am not sure about this statement, aren't we always cutting the corners to make things ~95% correct at scale to meet deadlines with our staffing/constraints?

Most of us, who doesnt work on Linux kernel, space shuttles, and near realtime OSes, we were writing good enough code to meet business requirements

My point is that coming up with the business requirements was always the hard part (unless you’re writing a scheduler)

> Stuffing agents somewhere they don't belong rather than making the system work better with the agents people already use.

I'm not bullish on LLM based agentic coding, but if there was ever a place to put an agent it would be in a centralised provider that has access to your CI, issues and source code. It seems like a perfect fit.

> if you make your build scripts bash, they’re Linux only

Git bash exists on windows and is perfectly usable.

It's still much less dependable compared to something fully supported like Bun.

I’m no bash lover, but I’ve used git bash as my primary driver on windows for I don’t know many years. It has its quirks (like any tool) but I don’t think I’d say it’s less dependable. Anecdotally I’ve had way more issues with node on windows than I have with git bash and node is fully supported.

Fully supported, until they sell it to someone.

I mean, Anthropic now own it

Yes, bash is the near-universal language for build scripts as long as git is the near-universal source control tool.

For day to day use, my base spec M1 MacBook Pro is snappier than my i9 desktop with 128GB of ram and a 4090.

My work laptop decided probably once a week to not go to sleep and just run its battery to 0.

My work PC will decide to not idle and will spin up fans arbitrarily in the evenings so I shut it down when I’m not using it.

> My work laptop decided probably once a week to not go to sleep and just run its battery to 0.

That reminds me that if Apple annoys me enough to switch back to linux at my main OS it will hurt on laptops :(

My Mac - couldn’t tell you, I just close the lid. My work laptop? Probably every day, as it makes its own mind up what it does when you close the lid. Even the “shut down” button in the start menu often restarts the machine in win 11.

My work desktop? Every day, and it takes > 30 seconds to go from off to desktop, and probably another minute or two for things like Docker to decide that they’ve actually started up.