nl

39 ms·

721.
by nl 9mo ago
> Much of these gains can be attributed to better tooling and harnesses around the models. This isn't the case. Take Claude Code and use it with Haiku, Sonnet and Opus. There's a huge difference in the capabilities of the model
722.
by nl 9mo ago
> When has any deal with China worked at a strategic advantage for the other side? Australia did well out of the Australian-China free trade agreement and then won the subsequent (Australian-caused) trade war. Turns out being able to pro
723.
by nl 9mo ago
I like Kimi-cli but it does leak memory. I was using it for multi-hour tasks scripted via an self-written orchestrator on a small VM and ended up switching away from it because it would run slower and slower over time.
724.
by nl 9mo ago
> I'll wait with bated breath for the millions of amazing apps which couldn't be coded before to start showing up Appstore releases were roughly linear until July 25 and are up 60% since then: https://www.coatue.com&
725.
by nl 9mo ago
Progress has not become linear. We've just hit the limits of what we can measure and explain easily. One year ago coding agents could barely do decent auto-complete. Now they can write whole applications. That's much more diffic
726.
by nl 9mo ago
Note that this is also a super interesting technique for specialising consumer facing apps like Lovable that need to generate code that matches your API very well. It's also a great approach for building custom languages.
727.
by nl 9mo ago
You are correct and the objection is just completely invalid. There's no way anyone would or should write tests like this at the client level. I think they are just arguing for the sake of arguing.
728.
by nl 9mo ago
I suspect the problem with VB is that VB 4 and 5 (which I think was that era) were so closely tied to the IDE it is difficult to work out what is going on without it. (I did Delphi back when VB6 was the other option so remember this problem
729.
by nl 9mo ago
it's very good at this BTW
730.
Agentic coding is accelerating app releases(coatue.com)
1 points by nl 9mo ago|0 comments
731.
by nl 9mo ago
Much of DID itself is basically a standardization of the idea behind Keybase, ie, using control of a private key as a marker of identity. This in itself is a pretty good idea (with some bad usability, but at least technically interesting) D
732.
by nl 9mo ago
This is fairly accurrate
733.
by nl 9mo ago
Both Apple and Android have had anti-stalking features for years.
734.
by nl 9mo ago
> why is a centralized “burn” able to completely prevent me from interacting with people using Bluesky? Presumably to stop credential reuse attacks on Bluesky itself? Bluesky is one instance and they should enforce security on that insta
735.
by nl 9mo ago
I think Codex is probably marginally stronger than Opus in my testing. But it's much much worse at writing issues than Claude models.
736.
by nl 9mo ago
Amp doesn't have a conventional model selector - you choose fast vs smart (I think that's what it is called). In smart mode it explores with Gemini Flash and writes with Opus. Opus is roughly the same speed as Codex, depending on
737.
by nl 9mo ago
Amp uses Gemini 3 Flash to explore code first. That's model is a great speed/intelligence trade-off especially for that use case.
738.
by nl 9mo ago
> hire a PhD to crack a tough problem for me, I don't need to go back and forth prompting him at a PhD level. I can set him loose on my problem and he'll come back to me with a solution. In my experience with many PhDs they are
739.
by nl 9mo ago
In general it will match the language style you use. If you ask a sophisticated question (lots of clauses, college reading level or above) it will respond in kind. You are basically moving where the generation happens in the latent space. B
740.
by nl 9mo ago
My question was more why a fork instead of doing the conventional "import tinygrad" into your own project. I don't think there is anywhere you are modifying tinygrad itself is there?
741.
Four Ingredients for Successful Retrofitting(bmin.ai)
2 points by nl 9mo ago|0 comments
742.
by nl 9mo ago
Why is this a fork of tinygrad and not just something that imports it?
743.
by nl 9mo ago
I just tried this on a production PR and I liked it. It found some things that Claude review missed, but missed some that Codex review found. I'd rank this the best of the three. Generally I actually like Gemini reviews a lot (I guess
744.
by nl 9mo ago
It certainly reduces the concern of PDF viewer vulnerability that allows local execution access which is the concern here.
745.
by nl 9mo ago
> Anthropic incorporated itself as a Delaware public-benefit corporation (PBC), which enables directors to balance stockholders' financial interests with its public benefit purpose. > Anthropic's "Long-Term Benefit Trus
746.
by nl 9mo ago
It's worth understanding the history of Anthropic. There's a lot of implied background that helps it make sense. To quote: > Founded by engineers who quit OpenAI due to tension over ethical and safety concerns, Anthropic has de
747.
by nl 9mo ago
Asynchronous tokens are significantly cheaper than ones where you need them immediately at a high rate. There a lots of free, high quality models on OpenRouter too, BTW...
748.
by nl 9mo ago
To quote you: > here is no RL for programming languages. and > Either RL works & you have evidence This is just so completely wrong, and here is the evidence. I think everyone in this thread is just surprised you don't seem t
749.
by nl 9mo ago
Typescript is a great type system for agents to use. It's expressive and the compiler is much faster than rust, so turn around is much quicker. I'm slowly accepting that Python's optional typing is mistake with AI agents, esp
750.
by nl 9mo ago
I use agentic coding daily and rarely write any code by hand. Here's what works for me: Spend a lot of time working out plans. If you have a feature, get Claude Opus to build a plan, then ask it "How many github issues should this

More ›