ggregoire

Is there a similar project for React? There are so much bad React code and wrong/old design patterns on the internet that surely ended in the training datasets

Are they? (genuine question, i've got no experience with Claude) This month I used 1 billion tokens (including cache) with DS4.1 flash and it cost me 7% of my $10 opencode go sub.

If you meant in developer time, then I don't know but it seems very fast to me (rarely thinks more than ~10sec, never more than ~30sec) and usually oneshots what I prompt it to do. I use opencode tui as harness.

Opencode go ($10), deepseek 4.1 flash high, opencode tui

I went directly with this setup, never tried claude/codex/cursor and their respective models.

When I order a Uber I just want a BYD… They should add the option in the app. Using the "Comfort" option increases the chance to get one, but I still get very bad/old cars occasionally.

Or maybe BYD could just start its own ridesharing service.

I'm sure a lot of people tried 1 model/harness and stuck with it without really comparing (same way I haven't changed my IDE and my terminal for the past 10 years), and eventually will try a Chinese model and be blown away.

I was discussing with a coworker yesterday who pays $200/month for cursor, has never tried any other harness and mostly use grok, in rare occasions opus 5.5. He seems to be regularly out of tokens. I told him I exclusively use DS 4.1 flash with opencode, spend like 100-150M tokens a day including cache and it oneshots all my tasks without issues. I barely reach 20% of my $10 sub at the end of the month. From his look I think he didn't believe me.

Meh… you can't create duplicated questions on SO so obviously 20 years later there is not much left to ask.

A better metric would be visits per month… which probably doesn't look very pretty too. I used to visit SO every day several times a day for like 15 years: ask a technical question on google, click on the first SO result, get the solution I was looking for. Now I still ask google but I just have to read the AI Overview to get the solution I'm looking for…

Yes, this can be abused to rewrite history too.

Related: Preserving Historical Truth: Detecting Historical Revisionism in Large Language Models

https://arxiv.org/html/2602.17433v1

"hacker" news… Even in the AI era, some people still enjoy understanding the technicalities of things

I think it was a great blog post.

The Adizero is incredibly popular among runners too

Might be a vocal minority. I used to visit SO every day, several times a day, for like 10+ years, got so much value from it and would not have changed anything to the site. "AI Overview" in Google completely killed it tho and I don't really see what they could do about it other than completely change the site into something else. RIP.

> On top of that, I enjoy programming, reverse engineering, etc. and I feel that the LLMs, while able to solve some problems or deliver some features, take that fun away.

Same, I prefer asking one or multiple very technical questions to Gemini, analyze, compare and understand the responses then implement it myself based on what I learned (or just integrate it to the codebase as it is, if I asked it to write a function) than delegating away all the fun to an agent.

> but the enthusiasm on the ground is lacking

Using claude and friends takes all the fun out of the job, so I'm not surprised engineers are not enthusiastic. It's cool for 1 month then you realize we went from solving problems and implementing algos and optimizing slow code and fixing security issues and other fun stuff, to writing prompts all day long.

Also, it's obviously faster to copy Slack 1-to-1 than inventing it from scratch. Making Slack was not just coding.

It's quite remarkable that a chrome extension can just update overnight and start injecting adware (or worse) and not a single warning from chrome. I shouldn't have to read hackernews to find out.

I've been playing exclusively CRPGs for the last 12 months or so, which was kinda a niche genre before the success of BG3. There are tons of way to beat those games and optimizing how you build your party and characters (what players call "min-maxing") while following a highly narrative story is a lot of fun. Most of them are quite old and often on sales for like 5 bucks on Steam, for which you get hundreds of hours of gameplay. A few recommendations: Obsidian's Pillars of Eternity 1 & 2, Owlcat's Pathfinders & W40K Rogue Trader, Larian's Divinity 1, 2 & BG3, Bioware's BG1 & BG2, etc…

> The difference isn't that the average techie doesn't dream of making a billion by any means necessary

That's actually the difference, most people don't want a billion

Any loser is a "full stack software engineer" nowadays thanks to claude.

That's a bit dishonest, the consensus on HN seems to be that LLMs are very good at oneshotting small projects from scratch. Especially when using super mainstream technologies like html and tailwind, as does the discussed website. And especially when it's a one time operation and the project will never need to be maintained, like the discussed website.

3. Coding is fun, prompting not so much

> I find Qwen useless for anything but coding tasks because if its insufferable sycophancy

We use Qwen at work since 2.0 for text/image/video analysis (summarization, categorization, NER, etc), I think it's impressive. We ask for JSON and always ask "do not explain your response".

> the moat of 'users won't be arsed to download a different app'???

don't even need to download anything, just open your browser and go to google.com to use gemini

last week-end, I've seen a non-tech friend who previously used chatGPT on his phone, just go on google to ask stuff to the AI (they have no idea it's gemini and it doesn't matter)

if you are not looking for having some kind of relationship with an AI (from what I understand people use chatGPT for this use case), but just looking for an AI to search stuff, then in my opinion you can't beat google search + gemini summary all at once for free with a single prompt

in 2 hours with claude code

> I don’t think people all over Europe/Asia/Africa migrate to China.

Learning mandarin is the major blocker imo, more people would move if the language was easier.

I use it in Google Search. For example yesterday I typed in Google "postgres generate series 24 hour" and this morning "ffmpeg convert mp4 to wav". Previously I would have clicked on the first StackOverflow result (RIP), now I just take it from the Gemini summary (I'd say 95% of the time it's correct for basic programming language questions. I remember some hallucinations about psycopg3 and date-fns tho. As usual with AI, you need to already know the answer, at least partially, to detect the bs).

Also what's great about Gemini in Google Search is that the answer comes with several links, I use them sometimes to validate the correctness of the solution, or check how old the solution is (I've never used chatGPT so I don't know if chatGPT does it).

> StackOverflow was well on its way to death even without ChatGPT, just look at the graph from [0]. It has been in steady consistent decline since 2014.

> [0] https://blog.pragmaticengineer.com/stack-overflow-is-almost-... (monthly question asked on Stack Overflow)

"monthly questions asked" is a weird metric to measure the decline of StackOverflow tho. How many times are people gonna ask how to compare 2 dates in python, or how to efficiently iterate an array in javascript? According to the duplicates rule on SO, should be once anyway. So it's just inevitable that "monthly questions asked" will forever decrease after reaching its peak, since everything has already been asked. Didn't mean it was dead tho, people still needed to visit the site to read the responses.

A better metric to measure its decline would be "monthly visits", which I guess was still pretty high pre LLM (100s of millions per month?), even if the "monthly questions asked" was declining. But now I imagine their "monthly visits" is closer to zero than 1M. I mean, even if you don't use Claude and its friends, searching anything about programming on Google returns a Gemini answer that probably comes from StackOverflow, removing any reason to ever visit the site…

Yes we agree, my comment was for tailwind users. :) I was replying to you because you gave the example of a Button component, and it's a good example to demonstrate that you don't need tailwind to style components.

> If everything in your code is a React component, I get why you would just want to write the styles right there.

Even for keeping the style close to the component, you can just use standard css.

Create a folder Button, create two files Button.tsx and Button.css in that folder, import the css file in the tsx file, add a class "button" on the first element the tsx file renders, start all the rules in the css file with ".button " to encapsulate the style.

People will say it's too much work, but it took me like 5 sec.

What infuriates me the most is iOS being absolutely unable to detect which language I'm currently writing in and automatically replacing words in one language to another. I write in 3 languages on a daily basis and it's making iOS totally lose its mind. For instance I'm 5 words deep into a message in French with a person I'm only speaking French with, and somehow iOS still thinks I'm making typos every other word and automatically replaces them with English or Spanish words.

And it wouldn't be so bad if moving the cursor at the end of a word or selecting a few letters in a word or even selecting an entire word wasn't nearly impossible on iOS (and I have relatively small fingers… I have no idea how people with big hands can do that stuff). Writing a 10 words messages can take me like 2 minutes sometimes because of all the errors made by iOS that I need to manually fix, and having to retry like 5 times to position the cursor successfully at the end of every word I need to rewrite in the correct language…

Do we know what model is used by Google Search to generate the AI summary?

I've noticed this week the AI summary now has a loader "Thinking…" (no idea if it was already there a few weeks ago). And after "Thinking…" it says "Searching…" and shows a list of favicons of popular websites (I guess it's generating the list of links on the right side of the AI summary?).

> How much you like coding. If you like coding then using AI is less fun.

I'm surprised this is never brought up here on "Hacker" News. I've been reading HN for 14 years and all this time I thought people here enjoyed programming. Turns out the majority hates it apparently.