rglover

Author here. I did this intentionally as I figured it would be the easiest way to communicate the idea to the broadest audience. 'Junior' is used in the sense of "lack of experience" here, not as a term of condescension.

OP here. Thanks for sharing!

Because that doesn't allow for sensationalizing the frustration.

I like LOEC: Lines of Effective Code.

Much easier to reason and make decisions if the bar is "yeah it exists, but is it effective relative to its intended purpose?" This weeds out buggy code as well as under performant code. If it's not effective (broken, too slow, etc), it either needs a refactor or thrown on the scrap pile.

[whoops]

Likely never as that's an ego death many of these guys couldn't handle. They'd implode from shame.

Always consider the source.

When someone rags on me or my work in a way I interpret as aggressive or dismissive, I make a point to go look at what they're up to. More often than not: nothing, or, something that screams "oh, your problem with me is despite you knowing things I don't, I also know things that you don't."

I agree with the general sentiment here, but it helps to mentally filter the passive aggression and condescension from the point they're making. More often than not they're correct, you just have to cut through the weeds to get to it. I'd say that's an occupational hazard as a programmer.

Good? No. But a reality (on the personality level) you have to deal with in this business. It sucks, but it is what it is.

I do deliveries via gig apps for extra cash and to get away from the computer and it's basically a complete toss up.

Some people are professional about it (I bring my own drink carriers, food bags, etc and have a minor Andretti gene so orders are as quick as the law allows), but most of the others I see doing it may as well not be. Slow, lazy, check nothing, and look like they just fell out of a methadone clinic (read: not the person you want showing up at your door in the dark at 8pm).

It's frustrating if you try to do it properly but get judged by customers like you're some out of work junkie (when, ironically, most people I deliver to live in a lower-income situation than I do).

From the driver's side, you get inundated with low-paying orders, often rapid-fire as you're driving to either pick up or deliver another order (like so bad that if you're not paying attention, you could get into an accident). Like literal Pavlov's dog, bell jingling over the top of the navigation instructions and a big pop up on screen. Essentially [1].

The entire thing is a psychological rat trap (by design as best as I can tell as these apps aren't terribly complicated to build/design; just scale), so it's not a total surprise when your driver fails to provide good service. They actually make it quite difficult to provide good service (e.g., I routinely get "already picked up" orders on my late night runs w/ no compensation for sitting/waiting on restaurant staff).

I think a lot of people should try doing deliveries via these apps a few times just to experience it. The psychological pressure alone likely makes a lot of people crack. It's all quite fucked up.

[1] https://www.youtube.com/watch?v=7my1G0TrwT8

Ah, disregard then.

Right, but does a business/company owe you the right to never require usage of AI to do your job (and by extension, owe you that job under those terms)?

Can't that be done without a performative "religious exemption"? You don't need that to say "yeah, I'm not using this stuff." I imagine in practice all this will accomplish is pissing people off who are already stressed about AI.

Can we please just look at LLMs like what they are and not some fantastical threat?

The fearmongering has cooked a lot of people's understanding of this stuff (and I wouldn't place the blame on them given the news they're likely consuming about AI).

"Play stupid games, win stupid prizes" is apt here. We don't need to return to mud huts and smoke signals because we can approximate intelligence.

An app for tracking monarchs with these tags is here for iOS [1].

[1] https://apps.apple.com/us/app/project-monarch-science/id6460...

Fear.

I live in the middle of nowhere and the local police have an MRAP. First time I saw it, I figured it was a display piece but nope, they use it for "SWAT" (in a heavily religious, sleepy, nothing but meth heads kind of low crime area). They've also littered ALPRs all over rural roads. I spot them constantly and they're always in strange "why there" places.

It's all ironic when you see the same people claiming "anti communism" and "low or no government" but then proceed to happily install the boot on people's faces. The cognitive dissonance is chefs kiss.

No.

Anecdotally, I remember far less frustration and far more enjoyment from software 15 years ago than I do today (what led me down the path to being a developer to begin with).

Today's software isn't even software. It's just a bunch of poorly made extraction machines backed by utter delusion that "it works" when a lot of it most certainly doesn't (or if it does, at best, it works as an efficient extraction machine and torture device with no real utility).

It seriously feels like we've been huffing gas for the last 15 years.

Calls too. My local cleaning business has been getting hammered with automated "your Google listing..." spam for weeks (virtual number w/ routing to our cells) now. Already hung up on 3-4 this morning/afternoon.

"Cheap" is up for debate, but the models themselves are great. Kimi K3 (and now 3.1 Pro) was the first open model I played with and didn't think to switch back to a frontier model. Now it's my daily driver and I have zero complaints.

The frontier labs are in a tough spot. It's hard to justify their cost when you can get something that's easily as good for cheaper (and anecdotally, more reliable/accessible). I do HITL so my perspective will be different than a fully-agentic setup, but as soon as I started using open stuff regularly, I realized that any moat the US labs had built is drying up.

Have we considered how things might improve if we try to be less ambitious?

You won't get magical productivity gains from a workforce heavily conditioned to following orders and a pre-packaged process (the Henry Ford model and its various bastardizations). In order to extract value from an LLM, you have to be able to think and ideate so that the inputs are of higher quality.

It should be of zero surprise that the rush into AI has yielded limited results (if we're honest, both in and outside of tech)—you won't get a bunch of artists or creative thinkers out of a pliant, ironically agentic labor force. That's why most people say "it doesn't work": it's because they lack the skills and knowledge to use it properly (but corporations want to delude themselves into thinking they have a genie in a lamp).

Using it now via API (3.1 Pro on Open Router) in a HITL setup (Cline in VSCode).

Off-hand it's quite good. I've been using K3 for the past few months almost exclusively for coding work and have been stunned by how good it is. Excited to see how far 3.1 Pro can be pushed.

Prednisone is nuts. I had to take that as a kid during Leukemia treatments and my mom would find me rifling through the kitchen cabinets at 7am eating my brother's single-serving pasta buckets on the floor like an animal.

> I mean is this not most of human existence through history though?

That's not a good thing. It means we're failing to progress as a species. If we keep repeating history (with now, unfettered access to it), we're devolving, not evolving.

> Nobody but me has it and it is great!

The zen of this position is severely underrated.

> What happens when each organization has its own entire stack though?

My personal opinion is that all hell breaks loose due to continuity issues, time and money waste (excessive token spend to simulate competency), and a general air of malaise to the point where problems are tolerated far longer than necessary because "the bot will get to it eventually."

> I get what you mean, but with how easy it is to get a gun (especially in red states)

Have you tried? I live in Tennessee (aggressively red) and own multiple firearms. Mandatory background checks and a mandatory 3-day wait [1] are the standard. Even online transfers from an FFL require submission of an ATF registration form.

Tangential, but it's not as easy as one would think. Also, there are a shocking number of posers down here. Stars and bars and "don't tread on me" but not a lot of commitment to "the mission." There are far more grown men driving lifted trucks with glow lights on them than serious threats.

[1] https://www.stateregstoday.com/politics/gun-control/firearm-...

> What do we have, no programmers, no one knowing what’s going on inside the test passing slopy black box, except a caretaker guy who’s kept their skills sharp by not using AI—with their time taken up trying to understand constantly rewritten code?

That's where we're headed but I'd say the smart teams will have that 'caretaker guy' use judgment to know when and when not to apply AI. That describes one of my businesses perfectly [1].

[1] https://lattice.engineering

It's already started imo. Spotify's desktop app on MacOS is becoming unusable (up to ~30s delays after clicking a button before the action runs—used to be snappy). I just saw the Epic Games launcher thing. Anecdotally, countless websites just break at random (more so than even when it was 100% human).