No argument that it works, it definitely does. I just think the LLMs have a hard time with tweaks/modifications of any kind, more so than what I see even in a custom GUI toolkit I’ve built that it has zero training data on.
Asking for simple changes to SwiftUI interfaces I’ve one-shot is always a frustrating experience for me
Logically this makes sense, but in practice it doesn’t, at least in my experience with SwiftUI. The LLM isn’t any better about generating/understanding it.
While the DSL is more formal than natural language, it’s not what we’re communicating to the LLM with, so it’s advantages are washed away. And typical code is more strict/rigorous than DSLs so I think that’s why I see worse results, because a typical languages compiler “catches” more mistakes, versus a DSL that’s easy to write but has lots of implicitness.
I’ve had the same journey experimenting with levels if abstraction too. Going lower, and exposing the LLM to the “full-stack” works much better than trying to build up abstractions it can’t see into without extra steps.
I don’t want to be too much of a hater, but these types of panacea/architecture posts are usually written by people who don’t work in the field, lack pressure or constraints, and get paid to goof around in castles of the mind. I would simply skip over it and hold my comments/opinions to myself, but they tend to have an outsized influence on software engineering practices.
> He says he writes slop code and doesn’t provide a single example or prove that statement.
You can look at the bun codebase. Moving fast has its drawbacks, and bun moved very fast. It’s not what I’d call idiomatic or exemplary Zig.
I think this is what Andrew was highlighting with “beginner energy”. Not knowing you shouldn’t rewrite a toolchain and take on an entire ecosystem can be a positive.
I think this is pretty common in business, where someone/something isn’t “great” but wins because they didn’t know any better and went headfirst into something experts avoided because they understand the challenges and what “doing it right” entails.
I tried CachyOS and Niri this weekend, and was pleasantly surprised hibernate/suspend actually worked with no customizations on my Desktop.
I’ve tried like 5 other distros and usually when it goes into sleep/hibernate the monitor never wakes up so I have to force shutdown using the power button.
I’m going to install this on an older machine and see how it goes. So far it’s been great!
> You talk about Anthropic giving the government 'enough rope' and getting 'owned', which seems like a tacit acknowledgement that the government does not give a shit about whatever the surface-level justification for this ban is.
I think you’re whitewashing Dario’s abhorrent behavior, and you think I’m whitewashing the government’s.
Unfortunately there’s no law against being a piece of shit and trying to pull the ladder up behind you, lobby for locking out your competitors under the guise of safety, etc..
So I have to settle for him getting swallowed by the morons he was influencing.
And if you think his concerns over safety were genuine I have a bridge to sell you.
Not sure why you're getting downvoted, or why the narrative here is all about this being payback for the Department of Whatever-the-fuck thing.
Dario has been spouting how his models are too dangerous, thinking he was playing 3D chess and got owned from my perspective. And there's the possiblility of insider plays by the current administration w/ OpenAI or SpaceX.
But Dario was running his own propaganda machine and gave them enough rope to do this.
Maybe just focusing on building solid models and running a business was the play, not trying for regulatory capture and being anti-competitive.
I skimmed the article, but couldn’t spot any details on their estimates. They mention 70b+ params as being large in several places. But we’ve had several 100b+ param models that trail Sonnet.
If Google is actually getting cheaper inference than everyone else with their TPUs, this smells like trouble to me. Maybe serving LLMs at a profit is proving difficult.
Or maybe they think because their benchmarks are good they can ramp up the prices. Seems like they don’t have the market share to justify a move like that yet to me.
The “Apps” app is so bad on macOS too (seems built off of Spotlight?). I’ll type the exact app name and it’ll suggest the one on my phone, an installer in Downloads, etc..
I think the goal was to do a massive rewrite for Anthropic (they acquired bun) and show that rewriting projects from lang -> lang with Claude can reduce security vulnerabilities to help with the hype for an IPO.
This is an interesting experiment but I’m skeptical of any claims of success by Jarred/Anthropic due to the incentive to hype agents. There’s probably a trillion dollars at stake with the IPO. And Anthropic seems to be developing this part of their business with Mythos and the super review features.
But I’d like to see the same experiment done on a project without so much relying on the story being success.
> 1) Higher level code is easier for LLMs to review and iterate upon. The more the intent is clear from the code, the easier it is for humans and LLMs to work with.
The counter-argument, and one that matches my experience is working at a lower level is actually beneficial for LLMs since they can see the whole picture and don’t have to guess at abstractions.
> it requires a hands-on approach and actually understanding what's being built.
I think this is true regardless of what language you’re using.
I’ve built a lot in Zig and there’s no difference between vibing stuff in it versus TypeScript/React. Claude can “one-shot” them both, and will mimic existing code or grep the standard library to figure everything out.
I think I’m building up an agentic IDE, just haven’t committed yet, but probably will this month.
One cool new thing I’m trying is running models directly w/ Vulkan. I’m about halfway there with my first model, but it’s going better/easier than I anticipated and I’m hoping I can make something very specialized and fast.
One would expect the platform owner (especially one where they own both the hardware AND software) to provide a reasonable / easy path to using LLMs if they are going to provide a framework for doing so. But Apple can’t because of how slow they ship updates
Asking for simple changes to SwiftUI interfaces I’ve one-shot is always a frustrating experience for me