orsorna

>Hey, I'm v1d0b0t Digital familiar. Builder of pipelines, breaker of specs.

I can't put my finger on it. Why is this writing style so embarrassing?

I think the more pressing issue is the medium degrading before the playback hardware. Disks have an average lifespan of 25 years. I surmise basic bluray hardware will last much longer.

> Some of these bots are sophisticated. They follow up in comments, respond to review feedback, and can follow intricate instructions. We require that servers pass validation checks on Glama, which involves signing up and configuring a Docker build. I know of at least one instance where a bot went through all of those steps. Impressive, honestly.

Impressive, but honestly meeting the bar. It's frankly disturbing that PRs are opened by agents and they often don't validate their changes. Almost all validations one might run don't even require inference!

Am I crazy? Do I take for granted that I:

- run local tests to catch regressions - run linting to catch code formatting and organization issues - verify CI build passes, which may include integration or live integration tests

Frankly these are /trivial/ tasks for an agent in 2026 to do. You'd expect a junior to fail at this and chastise a senior for skipping these. The fact that these agents don't perform these is a human operator failure.

It's not useful in a vacuum, but one of many degrees that can be combined to create a unique profile of you.

Well I just explained what I actually enjoy about programming, which is the results of it. Many jobs have intermediate boring steps that build to something satisfying.

>but why were you subjecting yourself to daily misery for so long in the first place? I don't get it.

It just meant it took a lot longer to build something, to get that satisfaction.

Your cancer cell analogy is moot unless you paint all AI generated applications to be unusable trash, which is not the case, and I wouldn't describe my own work with it. It's true that standards have dropped to the floor where anyone can "ship" something but doesn't mean it's good. I think I have a better handle on how to steer GenAI versus the average linkedinbro. But the divide between journey and destination is valid, I guess it's something that hasn't been explored until GenAI.

Well for one, programming actually sucks. Punching cards sucks. Copywriting sucks. Why? Well, implementation for the sake of implementation is nothing more than self-gratifying, and sole focus on it is an academic pursuit. The classic debate of which programming language is better is an argument of the best way to translate human ideas of logic into something that works. Sure programming is fun but I don't want to do it. What I do want to do is transform data or information into other kinds of information, and computing is a very, very convenient platform to do so, and programming allows manipulation of a substrate to perform such transformations.

I agree with OP because the journey itself rarely helps you focus on system architecture, deliverable products and how your downstream consumers use your product. And not just product in the commercial sense, but FOSS stuff or shareware I slap together because I want to share a solution to a problem with other people.

The gambling fallacy is tiresome as someone who, at least I believe, can question the bullshit models try to do sometimes. It is very much gambling for CEOs, idea men who do not have a technical floor to question model outputs.

If LLMs were /slow/ at getting a working product together combined with my human judgement, I wouldn't use them.

So, when I encounter someone who doesn't pin value into building something that performs useful work, only the actual journey of it, regardless of usefulness of said work, I take them as seriously as an old man playing with hobby trains. Not to disparage hobby trains, because model trains are awesome, but they are hubris.

But you're effectively asking a third party application, running in a browser no less (i already understand that a browser exposes WAY too much os level information), to query the OS for age information.

Someone better let Simon know!

Were you a kid then too? :)

Further, I don't argue that 100% prompting an application together isn't building something themselves. Built on the shoulders of leviathans, as libraries were built on the shoulders of giants of yore.

But an application that combines xyz features is novel in this scenario. There is inherent value in that.

I don't follow your analogy at all. Suppose I want to build an application with xyz features. My research yields that there are no such applications that include xyz features. However, there are plenty of applications that might have x feature, y feature or z feature, or a combination of two, but not all three.

If there are no such applications, I don't have a choice but to write it myself. This could take some time, especially if an MVP is all I'm interested in. LLMs are a novel tool in building an MVP. If time is a constraint, I can use an LLM, which should excel since xyz features are in its training set.

I suppose your analogy follows for developers who write applications that support abc features even though there are already applications out there that support abc features. Yes, I don't think that is very interesting. Your umpteenth clone of Snake is not interesting.

As someone who pulls a salary and does not get rewarded equity: agree!

Does the claw in the VM have proven capability (verified by your team) to track changes it makes to itself and persist across reboots? What about rollback capability?

It's difficult to describe how the US black community has uniquely suffered for centuries due to unique appeal of the mercantile class to the Catholic church about how it's actually okay to mistreat their human cargo and perform chattel slavery. Then, hundreds of years later the mercantile reason is forgotten, yet bigotry from it lays embedded in American society. And in other societies that took black people as slaves, to a lesser extent.

So a black segregated community does not sound very racist for these historical reasons. Not to say that desegregation in America has failed, but that there are infallible holdouts that wish to cling to a rotted out ideology that, again, its origins by its believers have largely forgotten.

Asian American communities are complicit, more or less unknowingly, to a lesser extent but only because they have aped to authority in America to achieve "whiteness".

The sad truth is that there is no community that prides itself on being white without parroting trickster mercantile talking points, of which again I remind that their origins are forgotten and they do not even know why they hate so fiercely.

So something like Blacksky is genuinely exciting in theory. Even though it will probably self select in a way that makes it inhospitable for new comers. Much like Bluesky...but, I'm describing a separate argument from historical bigotry, which I feel compelled to call out since its origins have--for the fourth time--been so forgotten.

So this wipes out the savings you'd obtain just by buying diapers by the bulk.

Too bad they didn't choose a more human interchange format...

No, I guess I meant editing to approach a text that doesn't look rushed over (LLM generation is a subset of such poor writings)

But you're right, it did hit the front page, and that says more about my sensibilities not lining up with whoever is voting the article up.

>Lean harder on your deterministic verification: unit tests, full stack tests, linters, formatters, static analysis

It's wild that the gamut of PRs being zipped around don't even do these. You would run such validations as a human...

What's wild is that with a few minutes of manual editing it would give exponential return. For instance, a lead sentence in your section saying "here's why X" that was already described by your subheading is unnecessary and could have been wholly removed.

The amount of effort put into this tool...just for hiring to your exact shop? I cannot imagine that's a good return on investment?

I've only ever stuck with urxvt and more recently foot since moving to Wayland. On Mac the default terminal works for me, though I've been using Ghostty to follow a trend, but I seriously don't understand the benefits netted to me? (other than the cool icon)

Seems like he ran out of money to get a good GPU though :)

Very cool! Where's the source so I can run this myself? :)

From what I can tell, the author (of the submission)* can afford being a multimillionaire buying thousand dollar displays and moving into extremely expensive detached housing, so his efforts don't apply to us breeder peon.

That being said, I see a lot of COTS products that fill this niche but all have privacy issues sadly.

You have spend tens of thousands of dollars on hardware to approach the reasoning and tool call levels of SOTA models...so, casually mentioning "just use local LLM" is out of reach for the common man.

I was thinking you could formally benchmark decks against each other enmasse. MTG is not my wheelhouse, but with YGO at least deck power is determined by frequency of use and placement at official tournaments. Imagine taking any permutation of cards, including undiscovered/untested ones, and simulating a vast amount of games in parallel.

Of course when you quantize deck quality to such a degree I'd argue it's not fun anymore. YGO is already not fun anymore because of this rampant quantization and it didn't even take LLMs to arrive here.

Was the project really ever valued that high? Seems like something that can be easily replicated and even properly thought out (re: pi). This guy just ran the social media hype train the right way.

Proof you can get hired off of a portfolio where you've never even viewed a single line of code form it. Definitely feel a mix of envy and admiration.

Claude, or any good agent, doesn't need MCP to do things. As long as it has access to a shell it can craft any command that it needs to fulfill its prompt.