People living in deindustrialized societies have no appreciation for how real world logistics work.

I don’t mean you turn the program itself into a genetic algorithm. I’m saying that the agentic loop for producing code acts as one. The code itself is just regular code. And the loop isn’t really any more inefficient than what you do as a developer. It almost never happens that you write perfect code on a first try in practice. You’ll write some code, run your tests, look how it did, and iterate. That’s precisely the same process the agent follows.

The difference from a typical genetic algorithm is that the LLM is not just randomly generating text that eventually fits into the shape you specified. It’s generating code that’s already close to what’s intended most of the time, and it just needs a bit of massaging to get completely right. That’s the feedback loop here.

I find I kind of look at the whole agentic harness setup as a genetic algorithm. Your tests and specs are the fitness function for the program you’re evolving, and the LLM is the mutator. At each step it generates some output, it gets tested against the fitness function, the LLM gets feedback and iterates on it. Eventually something working falls out in the end. The better you can define the selection criteria the more you box the agent in the better results you get.

The trick I can recommend for getting the model to code is to ask it to come up with a phased plan composed of focused features, and then to build each feature on its own branch. That way you have a clear unit of work that does a specific thing which makes it much easier to review the code. Can also recommend tools like https://github.com/Fission-AI/OpenSpec for making specs to box the model in when it works.

You can run the Gemma 4 and Qwen3.5 MoE models with as little as 12 GB of VRAM at 30-40 tps (Q4/Q5), and they both blow GPT-4o and DeepSeek R1 out of the water. But 64gb RAM is also not really out of scale with the cost of a shop tool in other trades. If you’re a professional that’s confident in a positive return on the investment, or just a hobbyist with the luxury budget for a “shop” that cost is well within consumer market. That’s not everybody, of course, but it’s not some inconceivable fantasy.

The key point is that local models continue to get more efficient and usable. You need high end consumer grade hardware today, but given how fast improvements are happening, it’s entirely likely that you’ll be able to get the same capability on even smaller hardware in a few months.

Indeed, I really hope Iran puts a stop to that finally.

We’ll definitely have to wait to see this all shakes out in practice, but I do think we’re starting to see material reality catch up with the empire. Iran proved to be too tough a nut to crack, and now the empire has no good options left. The fact that there are serious talks happening on Iranian terms shows that Iran is in the dominant position here.

the term is semantic drift

Actually, ReactOS and Wine have historically worked together and share significant technical overlap in the goal of reimplementing the Windows API, though they have different approaches and end goals. They’re separate projects now, but a lot of work in wine happened thanks to ReactOS.

yeah that’s true

And now everybody saw how the US can just cut access to their model at any time too.

And what’s preventing other countries from subsidizing their own essentials again?

Yeah, LaTex has been a pain to set up historically, so making it more accessible is very welcome.

that’s cause it’s literally what it is, legalized bribery

amazing

it sure does, the US is just the most blatant about it

Might also pave a path towards governments moving off windows if it can run existing software they use.

yeah the entitlement here is off the charts

If this turns out to be true, it might actually be a sign that Iran won the war and the US is settling on their terms.

Yeah, it’s something that could be done to recover oil in the long term, but doesn’t really help with the immediate problem in any way.

oh for sure