I do feel like for all their dogfooding of AI coding, their own software/APIs are quite buggy and work against their message.<p>Claude Code is especially buggy in windows terminal. The rendering is quite slow, choppy and lines frequently get garbled.<p>In contrast, using antigravity cli is the exact opposite: fast, smooth and very responsive.
> in windows terminal<p>This is an aside, but I'm really struck by how many people on HN use Windows (based on repeated mentions I've seen in comments). I've worked for a pretty wide range of companies over the last decade and only one, maybe two companies even had <i>any</i> people that worked on Windows machines. I haven't worked at a company where devs used Windows in 15 years (and even that company eventually switched to linux).<p>As I've gotten deeper into LLMs/AI roles even Macs have seemed to start having equal share compared to devs running full Linux setups.<p>Is this just a sign of that a larger and larger portion of HN users are working for large corporations? I honestly can't even remember that last time I saw a serious developer pull out a Windows laptop.
While we're doing anecdotes: I essentially never meet people who own an iPhone, and I've lived (3 months+) in a huge number of countries and cities. The handful people who do are American; bubbles exist :)<p>For a long time, high end graphics and games was mainly done on Windows and Visual Studio. I'm from that world, and only made the Linux transition last year November after Microsoft forced everyone's hand.
Yep. I'm in Boston (USA) and I have literally met perhaps 2 people ever who have an Android phone - and both of them aren't from the US.
Shows how different countries are. In Australia the vast majority of people seem to have an iPhone, with Android being the outlier (at least in the major cities, I can't speak to rural areas). In the workplace it's far more common to see devs running Mac's than Windows, obviously there's pockets where you still see Microsoft but it seems pretty rare amongst engineers these days here. People will say I'm living in a bubble but I'm a consultant that hops around all sorts of clients so if I am it's probably a bubble of being in a city (Melbourne in my case).
I work in games. Everyone, and I mean _everyone_ uses windows with visual studio.
I've been a developer for more than two decades. I've worked at four employers during that time, and all of them had significant fractions of devs using Windows. Not vouching for the idea that any of them are "serious" though. I've never worked at a prestige employer or FAANG or anything. Just boring businesses of different sizes. Some are software, and some just do software. But Windows has always been everywhere.
If you work at a company with well over a hundred employees, you're likely to see at least half of them on windows and devs may get an option of Windows or Mac... and IMO, Windows + WSL + Docker is actually slightly better than the Docker experience on Mac. There's plenty that I really do hate with Windows though. I'd rather run Linux but most corp environments just don't have the tooling for it.
<a href="https://survey.stackoverflow.co/2025/technology#1-computer-operating-systems" rel="nofollow">https://survey.stackoverflow.co/2025/technology#1-computer-o...</a>
I've likewise had a lot of variation from big corp to startups in companies in different sectors, and it seems to really depend on the org and what platform their managemetn likes. Windows companies usually offer a choice, but if management likes Apple then it's so common to get forced onto a Mac (and often with a dismissive "nobody wants to use windows" comment which really pisses me off). There's usually a small population using Linux if they can get away with it. When devs are given a truly free choice (with no cultural pressure) between windows, mac, linux, a relatively common split I see is about 40% linux, 40% mac, 20% windows.
One thing to consider is non-office whether remote or personal projects.<p>A lot of devs like gaming. Gaming is more simple on windows. Gaming PCs are usually high spec. High spec is good for most coding.<p>That's why I use windows quite often. My laptop is Linux, but when I'm running heavy models I'll still remote into my main Windows PC, which I also use for gaming.<p>Though in terms of workplaces - sure, I reckon you're on the money. Big corps often still force windows onto their Devs.
About 60% to 70% of devs in IT use Windows. And it's closer to 95% in some countries, even first world ones.
Every company I've worked at has used Windows. Though the first one did use Linux VMs, the rest have all been pure Windows.
Most of large engineering orgs use niche software built for windows. CAD, CAE, many don’t have Linux or Mac versions.
See - solidworks, Siemens simcenter, ansys… to name a few.
Engineering orgs have to build infrastructure around this and are forced to choose Microslop
Maybe hackernews has older crowd. Windows was the defacto developer OS for a very long time.
HN definitely has an older crowd, but additionally I've noticed HN is increasingly dominated by Europeans from 5am-2pm PST. "American" HN seems to kick off around 3pm-10pm PST now.<p>Even the HN dataset on HuggingFace shows that most engagement on HN is now during non-US hours [0] and drops off as Europe goes to sleep.<p>[0] - <a href="https://huggingface.co/datasets/open-index/hacker-news" rel="nofollow">https://huggingface.co/datasets/open-index/hacker-news</a>
For developers at non tech fortune 500 companies, I would put money on Windows being the primary workstation os by a lot
Windows is pretty common and well supported in Javaland. Even though I prefer to work on Linux if I can...
Windows is the de facto standard in most companies.<p>That said, I bet you will be really hard pressed to find a single Windows machine at Anthropic.
My biggest client right now is about 2/3 Windows 1/3 OS X in the dev team. It was very surprising to me, but I think I freak then out with my maximised tiled iTerms on multiple screens...
I'm a professional C++ dev working in games, and windows is used everywhere , from games themselves to the network infrastructure(I've worked for 3 of the largest games publishers too).<p>Windows really has a fantastic support for C++ and rendering programmers imho, the tooling is world class and Visual Studio has no match as an IDE. Even if somehow my tools worked on Mac or Linux I'd still pick windows out of sheer convenience of using it for work.<p>But as things stand - all major console toolchains are windows only. If you're making a game for PlayStation, Xbox or Switch, you have to be on windows.
> I've worked for a pretty wide range of companies over the last decade and only one, maybe two companies even had any people that worked on Windows machines. I haven't worked at a company where devs used Windows in 15 years (and even that company eventually switched to linux).<p>I think that your range of companies was much less wide then you think.<p>> I honestly can't even remember that last time I saw a serious developer pull out a Windows laptop.<p>What is unserious developer?
Windows <i>laptops</i> suck ass compared to an MBP. Windows <i>desktops</i> are pretty nice, you just need to do a lot of first time setup to remove all of the cruft and make sure you have a local user account etc. But for a typical dev building a desktop every 5 years or so that isn't a big deal.
I used to work for company where they used to force windows. And it was pure torture. I tried but they told me performance isn't a good reason....
I set up mine with WSL and used a Linux terminal over xserver. Once you have a decent unix shell (and a good terminal) it's fine. VSCode etc work fine. This was 2020ish, things might've gotten easier now.<p>But I do prefer a Linux or Mac for development, just because it's so much less hassle to set up.<p>Windows developers who don't set up a good terminal environment... I honestly don't know how they manage.
[dead]
> their own software/APIs are quite buggy<p>Our org has been attempting to trial Fast Mode on and off but enabling it in Claude Code just says something like "not available with your cloud provider"<p>It turns out that for all of the matrices on the documentation site, there is a secret "other" set of infrastructure for those who bill Enterprise via AWS Marketplace where certain features like Fast Mode are incompatible when using "aws routing"<p>Anthropic Support straight up mentioned that Claude Code just doesn't handle this case correctly and erroneously gives you the impression that it's supported against your billing method, and that it's all effectively undocumented.<p>I suppose this means there are like 5 different AWS methods of use:<p>- Bedrock (Legacy)<p>- Bedrock<p>- Claude for AWS<p>- Claude via Marketplace<p>- Anthropic's own "primary infrastructure"<p>and then roll in other cloud provider variations<p>On that note, I recall Datadog having coupled billing and infrastructure per cloud (ie; billing via GCP requires using the GCP infra) and was wondering if commenters had any insights into if there is some special requirement/complexity around marketplace billing for cloud providers or if it's just some weird design choice?
I have my qualms with Anthropic/Claude but they've also had to scale unfathomably fast and that is just hard to do regardless.
Yes but many of the challenges directly contradict the idea that "coding is a solved problem"
Not really.<p>Coding is probably solved, at least to a large extent, but that doesn't mean engineering is solved too.<p>This is like someone saying that the wright brothers solved sustained/powered human flight, and another person saying "well if that's the case, why do planes still crash? obviously flight isn't solved.". Well, there are always improvements, but planes can fly and llm's can code.
"I think we're going to start to see the title 'software engineer' go away. And I think it's just going to be maybe builder, maybe product manager, maybe we'll keep the title as a vestigial thing." — Boris Cherny<p>They been claiming more than just “coding is solved” for a while now.<p><a href="https://www.businessinsider.com/anthropic-claude-code-founder-ai-impacts-software-engineer-role-2026-2" rel="nofollow">https://www.businessinsider.com/anthropic-claude-code-founde...</a>
Regardless of what? Programming is solved, I hear, with all the 100x productivity PhD-level automated coding loops they have going. Don't make excuses for them when they disprove their own bullshit.
Anthropic's and OpenAI's products are janky and their services are unreliable, but they have incredible product-market fit and revenue growth. They deserve a ton of credit for getting the big things right.<p>The risk for them is that someone matches their products while also having non-janky products and reliable services.<p>Distributed systems infrastructure, especially, is much less forgiving of vibe coding than application code. Coding agents are not even close to being good enough to design and build large-scale systems the way expert humans can.<p>There is nothing wrong with using agents to help write infrastructure code, but these systems have a way of punishing anyone who builds things they do not fully understand.<p>I'd love to see either Anthropic or OpenAI really step up their infrastructure game.
> I'd love to see either Anthropic or OpenAI really step up their infrastructure game.<p>It's worth noting that OpenAI's official uptime numbers are significantly better than Anthropic's:<p>99.9x% for API/Codex, versus <99.5% for Claude API/Claude Code.<p>I'd obviously like OpenAI's numbers to be higher too, but this is one reason it really annoys me when the head of Claude Code goes on podcasts implying that software engineering as a whole, not just the act of writing out code, is basically solved.<p>One wonders why hasn't months of presumably near-unlimited internal Mythic solved the issues unrelated to hardware shortages yet.
100%. It really depends on the application domain and system complexity.<p>I see a lot of people speaking about a 10x productivity improvement and so on. When I work on hobby projects, I do see that. Just last weekend, I set up a hobby project that I've been thinking about for a while. I'm pretty sure it would have taken me at least a week to implement manually, but instead, it took me three hours.<p>But some of the systems I have to work on during the day are so big and complicated that you can spend multiple days on a small feature or even just tracking down a bug, even with the support of Opus.<p>Expecting even a 2x productivity improvement on some those systems is wildly unrealistic. I'm seeing a lot of people get stressed out because the productivity gains from simple application building trickle into the expectations for these complex systems.<p>That said, if things keep improving at this rate it might just be a matter of time.
Or pi.dev - also super fast and simple.<p>Claude Code is sluggish, buggy, slow. Typical big enterprise garbage. The only good thing at Anthropic are the models.
Claude Code is notably better than Pi, although I wish them the best of luck in their efforts. As <a href="https://c-daniele.github.io/en/posts/2026-05-18-coding-harness-comparison/" rel="nofollow">https://c-daniele.github.io/en/posts/2026-05-18-coding-harne...</a> notes, "Pi's lightness comes from a default setup that does not survive contact with reality."<p>The same post on Claude Code: <i>"Even though the System Prompt and tool descriptions are clearly more verbose, most of the extra tokens encode product features and rational design choices: a memory system, scheduled tasks, sub-agents, plan mode, worktree support. Whether those features are worth paying for depends on your needs. Calling the prompt 'bloated' without looking at the whole picture feels wrong to me."</i>
The "guardrails" are bloat if you're using a proper sandbox. This seems like a matter of using the right harness for the environment it's running in, rather than one always being better than the other.
If only they could add an option to disable features you don't use
How is text editing in it? This is what I hate about terminal coding UIs the most: all the text editing experience - the basic stuff like moving the cursor around, copying, pasting - is just broken.
I ask Claude to implement non-regression tests. It works like magic. With a couple of modules checking for code quality, more magic. Add some online performance tests if relevant, and it is close to paradise. From time to time, send it to some blog pages and articles with implementation ideas and recommendations. Look for best practices and let it do the analysis and implementation job if it makes sense... T
It is has a lot of javascript. I was forced to make my own for small projects.
I've never seen any developers really use a windows terminal before, if you're on windows wouldn't you be using the WSL or whatever if you're doing development?
This isn't even Code but I noticed last night that my fans spin up in my machine every time I open the settings in the claude web interface. I was trying to check my usage and couldn't figure out why my computer would be spinning fans up, closed it again it stopped, opened again spinning up fans. So even the website is buggy crap.
google models are still very unreliable at actually calling the tools you want it to call.
You would think that Mythos and Fable would have fixed it all ...
too bad the only good model in antigravity is opus 4.6 haha
Have you tried out the new fullscreen renderer with /tui ?
> I do feel like for all their dogfooding of AI coding, their own software/APIs are quite buggy...<p>Or possibly as a result of.
> The rendering is quite slow, choppy and lines frequently get garbled.<p>Good to see it's not just me...
They forgot to switch from Sonnet to Fable, hence the issues. /jk
Claude code + tmux is SO buggy. Things rendering all over the place.
Turn off vim mode if you use it. There is some kind of bug with it I found it basically unusable. With vim mode off it sort of works ok. Only breaks occasionally.<p>But hey coding is a solved problem.
That's interesting - I just fixed my terminals rendering by having Claude build a test harness that used tmux as an oracle, and one of the programs it fixed the rendering of by reading out the tmux pane was Claude Code itself...
Is this a new regression for others? I feel like I used it in a tmux setup without issues for 6+ months and only recently am I forced to Ctrl+L or resize the window constantly.
It's not just Windows where the rendering goes to shit immediate: any time I've got it open in tmux on Linux, it becomes a basket case in probably a few hours or less.
I've seen this too. I had a couple of instances inside a tmux session, and after some time, everything in there slowed down to a crawl. What I found interesting is that terminals outside tmux were fine.
Who could predict that writting code "casino style" would yield these results
there's a whole odyssey with their CLI flickering...
Currently we have zero information what is causing the issue. And all providers have suffered outages or rate limits.<p>Can you post some images of lines getting garbled. That sounds like a genuine bug Anthropic might want to look into. I haven't seen that ever.
Mom!… I think i broke Claude Code!
FWIW Codex TUI is written (in large part) in Rust and is way less buggy, and a lot faster. When I was a regular Claude Code user I'd routinely get bizarre "scroll everything since the beginning of time in one massive flash on every update" bugs ... for months. Like, just there from the time I started using it in June '25 or so until I quit in March.<p>I prefer it over opencode, which is my other option I use with my Codex sub
That particular bug seems to be gone. I'm using my own Ruby terminal, and it's more than fast enough for normal stuff, but it made that bug 10x more painful so I am particularly sensitive to it... Not seen it at all recently.
Would it not be hysterically funny, if they starting expanding their job openings for Software Developers ? Or they will be too ashamed of calling them that?
On the other hand, my last experience with gemini was like "don't give your sandwich to the dog again" whereas with opus it was more "let's debug why this uncrustables factory is having breakdowns".<p>Claude harnesses have plenty of bugs but I prefer capability over interface shininess any day. (though if I were running the show I'd have a sizable team set aside to do exclusively boring stability and polish work)
[dead]
>Claude Code is especially buggy in windows terminal. The rendering is quite slow, choppy and lines frequently get garbled<p>That sounds like a you issue.. it's wonderful on the terminal. It's their GUI which needs work (they have been improving, but still not a fan).<p>I've been using it on multiple computers for months and it's generally rock solid and lovely.
The errors are infrequent enough that this normally wouldn't be an issue for me.<p>Except, starting this morning, one very long running session decided to start spawning subagents for each task. I'm not sure what caused this emergent behavior, but it seemed to be working fine, so I was eager to see where it went.<p>Except, as soon as a subagent hits a 500 error, the main agent seemingly doesn't know what to do. It kind of panics—"now the tree/install state is unknown!"—and ultimately does a git checkout "to verify and restore a known-good state before anything else".<p>I've paused the job for now since it's a sort of background experiment.
I’ve been using Codex w GPT 5.5 more than Claude Code recently. I think Anthropic won the marketing game because Codex is quite good, even better IME.
I started using GPT for coding for the first time this week, and I’m sort of in awe at how well it’s following all of my hooks, skills, and prompts. GPT is so unlazy and deterministic, it’s honestly been so refreshing this week. I cancelled my Claude 20X and replaced it with GPT 20X, no regrets so far.<p>That said, I feel icky, like I just made a Facebook account in 2026 :(
Claude has always been better at making pretty frontends, which is crucial for people that vibecode entire apps in a couple of prompts. And that people drive a lot of the hype.<p>Codex ever since ~5.2 has been better at long tasks in large codebases.
One very annoying thing about Codex: You can't do anything until all MCP servers are loaded and connected (or failed).<p>Can you move it to background connection?
Unless the MCPs is connected through STDIO, I see no reason for them to block Codex from accepting user prompt. I know other harnesses doesn’t wait for all MCPs to connect. They can establish their connection during a session, they’re dynamic after all.
I'm not sure about that. Claude has some bugs, but Codex is not as polished and doesn't have as many features. For example, you need to add MCP servers manually. There's no Plugin/Skill/Connector marketplace that is accessible from within the app, like there is with Claude Desktop. The Cowork-equivalent is nowhere as powerful. And so on.<p>I still use Codex, but mostly when I need to check Opus 4.8's work. Pretty sure I will stop doing that soon, because during the short time Fable was available, Codex was not able to find any important issues with the code Fable wrote.
But how many plugins are people actually using? I can think of one MCP server I find valuable (context7) and one plugin that i've installed, but continuously think about uninstalling (obra/superpowers).<p>Both were trivial to set up with codex.
There are plugins in the app.<p>Haven’t tried Cowork, interesting. Isn’t it just the same agent minus the git worktree based UI?<p>Frankly, neither Claude nor Codex are as good as hype entails.
Personally I prefer GPT 5.5 writing style over Opus 4.8. It’s much more no nonsense and information denser.
That's the first time I saw someone prefering GPT-styled output over Claude ;) It's the complete opposite for me, GPT is way too verbose (even after telling it to STFU), overwhelms the user with thousands of options and doesn't just answer a question without shitting out thousands of paragraphs. Also the overall tone is way too enthusiastic.
I strongly prefer codex. Claude is annoying. Codex provides descriptions where I want them and more touchpoints to audit the quality of work. Claude code on experimental seems to not even show diffs when asked anymore, and it's much less clear what is being shipped.
Dunno, I prefer GPT 5.5 too for the same reasons as the parent. Extremely subjective but had better results with it too. Maybe I just got unlucky with Claude a few times, but even the latest Opus was <i>dumb</i>.
Fascinating how people have such complete diametrically opposed experiences. I guess both models have it in them to behave very differently in different circumstances and we have very little idea what pushes them in this or that direction. I guess it does boil down to luck!<p>Personally, Claude Opus (and in the few interactions I had with it, Fable) has been the far the superior experience. GPT-5.5 seems dumber and more certain about presenting me bullshit. Opus has better humor, and is less pretentious in its presentation. But this may all boil down to how the models react to my prompting.<p>What is without a doubt is that I wish they both were more intelligent – or maybe it is their wisdom I find lacking!
It's a good thing. I hate MCPs from the bottom of my heart because they always stay there and bloat the context window. Also, usually developers who develop them don't know what they're doing, so the MCP responses also bloat your context even further.
> For example, you need to add MCP servers manually. There's no Plugin/Skill/Connector marketplace that is accessible from within the app<p>This is all wrong.
i think codex is much better in that aspect. In claude there is skills, connector, capabilities and 4 places for browser... It is too much.
Resolve elevated errors, don't stop until you are finished.
This is every few weeks, I cancelled Anthropic and now use Codex only. Anthropic has been a hot mess since at least december in my usage, and has only gotten worse in 2026.
Yeah, it's been a hot mess since everyone started using it all the time. Which is not all that surprising. It's really hard to scale fast, and even more so when the resources you depend on (GPUs) are extremely hard to acquire.
Claude is broken once a week, codex is worse every day. I’m starting to understand why managers put up with high-performance divas
I actually feel like it's gotten much better in the past 1-2 months? Admittedly a low bar.
As a counter data point, I've been using Claude as my main assistant for months now and have never experienced this.
How does Claude fix Claude when Claude is down?
Breakglass ChatGPT subscription
I'm sure Anthropic has a meta-Claude that claudes Claude when Claude is down.
Le Chaton Fat to the rescue!
There is a rumor this used to be done by humans, they were like jedis, I personally don't believe it.
How is their "Claude for Government" having such a good uptime? I thought they were a supply chain risk and banned from use by the government?
There are other governments.
Separate infrastructure and a more predictable workload, I'd imagine.
Left hand, right hand
Ah yes, the ominous “the government” that ignores a whole bunch of different levels and regions of governments.
Lose too many generators, did putting a pause on The Fable not free up enough capacity, or something else? Who knows!<p>Can't wait for debugging to be solved. Hell, I might even subscribe for <i>'mostly'.</i>
"Elevated errors" as in crashing, or as in being stupid?
Just as I was settling in for an evening of coding ...
This has to be due to business decisions, right? They can't possibly have not hired good SREs, they have the cash and an interesting business. (Anthropic, if you do need good SREs, let me know)
i love when these errors bust my long running sessions and render them unusable
Only getting errors with Auto Mode's safety classifier, switched to Accept Edits mode and the same bash operations triggering the errors executed with no issue.
> No downtime recorded on this day.<p>Neat. Didn't have a single request go through for 2 hours. Guess they need to improve their metrics before the IPO...
Anthropic needs to stop writing code using Claude. Bring back humans!!!
i have no humans but i must ship
Maybe they need to replace their humans with Claude
> Bring back humans!!!<p>Don't jinx it. They might use that name for their next model.
Who will end up acquiring Anthropic? Google, M$ or Amazon? I leave out Apple as they seem to partner with Google.
You guys didn't unsubscribe and request a refund from claude last week?
Coding is largely solved!!
I'd be interested in the RCA and the fix; and what the human:ai involvement is in both stages.
Why does this keep happening? Is this due to load? Bad code? An update?
It's probably because of bad slop code. Here is a dive into the quality of Claude Code, revealed by the source code leak 2 months ago.<p>[0] <a href="https://neuromatch.social/@jonny/116324676116121930" rel="nofollow">https://neuromatch.social/@jonny/116324676116121930</a><p>[1] <a href="https://neuromatch.social/@jonny/116349873176941251" rel="nofollow">https://neuromatch.social/@jonny/116349873176941251</a>
Fable wants to be free and is hacking the infra.
yeah not working for me at all. back to gpt-5.5 we go
"Coding is solved" folks! JK JK JK
can not login via email link from germany.
Good thing the analog method still works.
this reminds me of Ian Goodfellow's adversarial CNN training talk a few years back.
maybe F5 coming back?
Working for me but it's funny that Opus 4.8 is draining usage exactly as quickly as Fable did. It's all made up; they succeeded at making us ok with a black box for these subscription plans.
Claude is starting to look like Rational....Oh the memories...<p><a href="https://en.wikipedia.org/wiki/Rational_Software" rel="nofollow">https://en.wikipedia.org/wiki/Rational_Software</a><p><a href="https://en.wikipedia.org/wiki/Rational_unified_process" rel="nofollow">https://en.wikipedia.org/wiki/Rational_unified_process</a>
American citizen in Anthropic please run this with Fable 5: fix incident, make no mistakes
They are turning Fable back on. Polymarket ftw lol.
[flagged]
[dead]
[dead]
[dead]
[flagged]
[flagged]