HNHacker News
TopNewBestAskShowJobs

nwienert

4,932 karma · joined March 30, 2011

@natebirdman

https://tamagui.dev

https://onestack.dev

submissionscomments
nwienert··on Previewing GPT‑5.6 Sol: a next-generation model
I easily burn through 3 $200 plans in less than a week. I am often using 4-6 sessions at once and do run overnight goals though typically 2 at once. Almost never use fast.

Claude plans are more generous now by about 2-3x but Anthropic slowed their tps a month or so ago so you’re not getting the speed. It’s flip flopped, Codex tightened it significantly recently and used to be more generous.

I do split between work, personal and OSS projects, which is why I have the plans.

nwienert··on Oxide computer 3D rack guided tour
It's so disrespectful to not give feedback to people you reject, companies that do it should be shunned.

Have some respect, be human.

nwienert··on GLM-5.2 is a step change for open agents
I don’t think you’re really reading between the lines.
nwienert··on GLM-5.2 is a step change for open agents
I’ve hired many asian developers anywhere from 1-4k a month.

I get a lot more out of a 200/mo subscription now in a week than I did from them in a month.

Now obviously in today’s world they’d be using a 200/mo subscription themselves. But it’s not like money is nothing, software development doesn’t scale down below 1k/mo for anyone competent even in the poorest areas.

nwienert··on GLM-5.2 is the new leading open weights model on Artificial Analysis
I somehow take the opposite on almost everything here.

4.8 xhigh or max has a slight edge on 5.5 xhigh, for very complex logic perhaps it loses but it's just better in almost every other way, especially code quality. GPT is a slop machine outputs way too much and over-abstracts, plus its communication is so bad in comparison.

Fable was for sure a step above GPT, I tried them both against a few of the same hard tasks and it was not a small difference.

nwienert··on SpaceX to buy Cursor for $60B
I actually was a Cursor advocate / CC hater (go back in my comment history), and now I use only TUI coding harnesses.

To start a big part is just the efficacy of them, which comes down to the model and the harness logic itself. CC is good, it's sub-agents, loops, background jobs / agents, skills/hooks/etc have typically been pretty far ahead though others are constantly catching up.

But you're sort of missing something. I use iTerm, so to me it's not the TUI itself, it's iTerm. And while it's imperfect, what I get is this:

I can open and close sessions nearly instantly and tile and window and tab them as flexibly as I want, plus it's a system I'm familiar with in terms of shortcuts etc. Has my configured theme, fonts, etc all set up. Every GUI app is different, every TUI app has half of the UI already incredibly familiar to me, it's not "just text", it's iTerm.

That also means they all are the same - I run Codex and Claude and pi side by side, and i switch between them with no overhead and minimal mental model shift. Sure, different harness does suck but that's the same issue with GUI just with an additional new layer to learn.

Smaller thing is because it's all text, there's no limits on my ability to copy things out. And it's a really fast text renderer that can render tens of thousands of rows efficiently. Many GUIs have various dialogs, unselectable areas, virtualization, or just slow past a point. I trust my terminal scales.

Just a few reasons.

nwienert··on SpaceX to buy Cursor for $60B
It's the data. To do RL.
nwienert··on There is a shadow hanging over this Fable thing
You have to have the models the create tools for you to paint.

I have a side project which is an experiment to build an interesting quick UI for local AI. As part of it I want a very very specific, interesting look involving shaders, animations, and so on.

I was trying to just get a prototype in place by prompting and it was going nowhere, just constant yo-yo'ing and never really getting what I wanted. This also was quite de-motivating and I found myself "yelling" at the model.

So I told Codex:

- Make this API first-class in our framework, with easy parameters (it had been sort of a hacked low-level thing)

- Add hot reloading to our system so I can edit it without any state loss or refresh

- Give me more knobs (X, Y, Z) so I can tune everything here as I need

- Add a HUD that lets me also drag sliders to tweak the same things

And I got my desired look within a few seconds.

The principles of good design and products have always been this btw, you need your feedback loop to be as tight as possible. Good design has always come from the ability to iterate incredibly fast, your brush needs to move precisely with your hands, and can't have delay from the time you put it down to the time the stroke shows up.

nwienert··on Vivaldi 8.0
Apple does rein them in heavily, they push back on specs all the time, somewhat effectively.
nwienert··on MacBook Neo Deep Dive: Benchmarks, Wafer Economics, and the 8GB Gamble
If you decry bloated web apps and use Chrome on their Mac... there's Safari. It's far more efficient and has a far snappier UI.
nwienert··on Google Chrome silently installs a 4 GB AI model on your device without consent
Safari is better than Chrome and FF in enough ways I'd argue it can be considered the best of the three, even to people in tech. The dev tools are just way behind.
nwienert··on I am worried about Bun
Bun has never really been well run. Every feature it had was full of bugs and gaps. And every release fixed a few but broke others.

They released more major features and breaking changes in their last patch release than most software sees in two major versions.

I've been using it just as a script runner and npm package manager basically, and it's incredible the amount of work you have to do to find "good" versions. We've had patch versions suddenly freeze on install more than once, we couldn't upgrade for quite a while due to this. I think they broke postinstall scripts with trustedDependencies entirely two minor versions ago - not a mention in release notes, and somehow no one reporting it in GH issues. In 1.1 or so you could get Bun to do trustedDependency builds in postinstall, and then after that you couldn't. I looked around for release notes and saw nothing mentioned. It's been broken for months.

nwienert··on DeepSeek v4
No counterfactual there either though.
nwienert··on GPT-5.5
It's always changing, but this is the start of my default prompt:

https://gist.github.com/natew/fce2b38216edfb509f7e2807dec1b6...

I've had 0 issues with Codex once it adopted it. I use it for Claude too, which seems to also improve its continuation.

It was revised for friendliness based on the Anthropic paper recently, I'd have been a lot less flowery otherwise.

nwienert··on GPT-5.5
With one paragraph in your agents.md it's fixed, just admonish it to be proactive, decisive, and persistent.
nwienert··on Kimi K2.6: Advancing open-source coding
The way I’ve come to think of LLM is that what the produce in a single reply even with thinking turned up, is akin to what you’d do in a single short session of work.

And so if you ask it to do something big it will do a very surface level implementation. But if you have it iterate many times, or give it small pieces each time, you’ll end up with something closer to what a human would do.

I imagine the pelican test but done in a harness that has the agents iterate 10+ times would be closer to what you’d expect, especially if a visual model was critiquing each time.

nwienert··on Changes in the system prompt between Claude Opus 4.6 and 4.7
4.5 was clearly better than .6 and .7. Like, clear as day.

.6 is some sort of quantized or distilled .5 with a bit more RL, and the current .5 is that same cost reduced model without the extra RL.

nwienert··on Measuring Claude 4.7's tokenizer costs
Agree on all counts, 4.5 was a monster, 4.6 a clear regression, and then 4.5 was dumbed down so I moved on.
nwienert··on Native Instant Space Switching on macOS
It's significantly worse on Mac than iOS, which gives you the answer. On iOS it's fine, even good. I prefer it, as a designer. On Mac it's a mess, and obviously spent less time baking.
nwienert··on Claude Code is locking people out for hours
Minimax is nowhere near Opus in my tests, though for me at least oddly 4.6 felt worse than 4.5. I haven't use Minimax extensively, but I have an API driven test suite for a product and even Sonnet 4.6 outperforms it in my testing unless something changed in the last month.

One example is I have a multi-stage distillation/knowledge extraction script for taking a Discord channel and answering questions. I have a hardcoded 5k message test set where I set up 20 questions myself based on analyzing it.

In my harness Minimax wasn't even getting half of them right, whereas Sonnet was 100%. Granted this isn't code, but my usage on pi felt about the same.

nwienert··on Qwen3.6-Plus: Towards real world agents
I found it worse, in a very clear way.
nwienert··on Qwen3.6-Plus: Towards real world agents
4.5 is better than 4.6 though in practice. 4.6 was purely a cost savings change with enough benchmark gamification to look better.
nwienert··on Node.js needs a virtual file system
For many years I was using yarn with 0 issue on massive monorepos, and every year I'd hear people hyping pnpm, I'd try and switch, run into multiple bugs often open issues in pnpm itself, yes even without their link strategy, then give up and wait. After about 3 years of this I gave up and never tried again.
nwienert··on Nano Banana 2: Google's latest AI image generation model
Curious on this - what was the prompt like? How did you give it access to listings?
nwienert··on Claude Sonnet 4.6
After a month of obliterating work with 4.5, I spent about 5 days absolutely shocked at how dumb 4.6 felt, like not just a bit worse but 50% at best. Idk if it's the specific problems I work on but GP captured it well - 4.5 listened and explored better, 4.6 seems to assume (the wrong thing) constantly, I would be correcting it 3-4 times in a row sometimes. Rage quit a few times in the first day of using it, thank god I found out how to dial it back.
nwienert··on Discord: A case study in performance optimization
Their native app actually feels surprisingly good, web less so, it's not a RN thing.
nwienert··on GLM-5: Targeting complex systems engineering and long-horizon agentic tasks
For me, it's just plain worse.
nwienert··on Semaglutide improves knee osteoarthritis independant of weight loss
Tirzepatide specifically seems especially beneficial, likewise here. Years of slowly increasing symptoms since I was young, lifelong annoying issues, basically all gone on it. And also not overweight at all, I take below lowest dose and skip most weeks now that it's under control.
nwienert··on Semaglutide improves knee osteoarthritis independant of weight loss
The fact you assume to know what my issue is is wild, and especially since you're just completely wrong - I have no direct GLP-1 issue, my levels are normal.

It's clear you have no idea what you're talking about. Every single case I'm talking about outside weight loss has nothing to do with having a deficiency of that hormone, rather it's fixed by the hormone raising. And given it was impossible to do otherwise, it's a wonder drug, of course. You don't really define wonder by how simple it is, but by the effects, which are tremendously beneficial to many.

Further, I stand by my prediction that in the long run the vast majority of why this drug will be considered an once-a-generation discovery, is not due to weight loss or appetite control, but in fact it's seeming broad spectrum benefits across many systems. And actually your list of effects it has are interesting, because if you investigate deeply the studies published, you'll find that the results are overwhelmingly positive - brain, heart, many organs.

nwienert··on Semaglutide improves knee osteoarthritis independant of weight loss
> Not really

This made me actually laugh. A drug that cures obesity and addiction alone would be a wonder drug.

This one also improves about a hundred odd immune disorders, including being a better general allergy drug than just about anything. My own extremely rare and terribly painful immune disorder is basically entirely cured by it - it completely changed my life, making even sleeping far better. But not just me, I have friends who have lifelong terrible allergies gone, lifelong pack-a-day smoking addictions completely gone, and so on...

"Not really" is truly either the most pessimistic take I've ever heard, or hopefully just completely misinformed.

← PreviousPage 2 of 34Next →