I often break these rules for one specific aspect of my personal projects: if it has a web frontend, I don't want to know what kind of CSS magic the agent used to make it look as it looks. I'll happily accept whatever unmaintainable AI slop it produces, because I don't want to spend any time figuring out if my understanding of flexbox (or anything else related) is wrong again, and why.
Today I learned (thanks to this article) that I can use timg to display images right in my standard macOS terminal, even without switching to kitty or any other fancy thing. Not pixel perfect of course, but still, much faster to go through icons or other pictures than opening in a separate Preview window. A simple "brew install timg" worked for me. Will surely save me some clicks!
Which is a very similar approach to any serious code. If you just hired a very clever, enormously knowledgeable intern, and they wrote a bunch of code for you overnight, you would probably review it.
Yes, in some cases, either hobby projects or throwaway code, you could just take it and use it as is, and I surely do, for the code no one cares about. But at work, I would rather review it.
Windows XP was released when I was in college, and I remember discussing this picture with my friends guessing where it could've been taken. Never thought about California back then, let alone moving there.
A little bit less than twenty years later, few years after moving to the US, my family and I were driving somewhere near Sonoma, CA, enjoying the views, and someone in the car said something like that this looks like that Windows background. Quick check with Google, and sure enough, we were less than a mile away. We didn't stop, but surely got some photos.
It's more like ∞x (or N/Ax if you prefer) because the majority of the projects I did with LLM agents wouldn't have existed without them, because I would've never found enough time to work on them.
One of the latest things I made with Claude was a tool that allowed me to move a bunch of very low traffic Cloud Run services to a single VPS without losing any of the Cloud Run benefits such as easy Docker-based deployment and automatic certificate provisioning. I thought about making something like that for quite some time, and Claude finally made it possible, which makes me quite happy.
The fun thing here is that no other soul genuinely cares about it, or any other code I might publish. The code, especially AI generated, is so cheap that if anyone wants to repeat my steps to get rid of Cloud Run services, they will probably vibe-code their own tool instead of figuring out how to use mine, just like I did that instead of spending time on learning Dokku or similar solutions.
So, yes, 10x and more, but no one cares about the result, which makes the whole 10x measurement less useful.
There have been plenty of libraries/tools that got people from "I would like to do foo" to "oh, this tool makes foo possible in under 10 hours, I should start working on foo" in the past. LLMs have done this for many foos, which is great.
But I'm with hansvm - I haven't actually seen anyone plausibly maintain 10x. 10x is different from getting people past their activation cost.
The incredulity at 10x claims is often unearned because how much do these skeptics actually notice and appreciate the depth of work of ten developers collaborating on something (if not their own org)? Dev output slips by quietly. There are reams of unnoticed projects even at the scale of a life’s work.
I'm assuming the devs are seeing 10x code generation and equating that to the improvement.
It's when they practically ignore the rabbit holes where it's suspect. I'm definitely seeing speed ups. I troubleshot a linux system yesterday with minimal effort using a local llm. It likely would have taken me a few hours to locate all the docs & testing procedures. the llm did it with only a few prompts. To ensure it did it correctly, I had to interrogate it a few times before letting it proceed.
Humans make really bad scientists, and it takes a lot of effort to properly catalog and provide statistics for these things.
There is an improvement, but I doubt any random dev can give a real estimate since before LLMs they couldnt really give you a real estimate anyway. I do know when I encounter a bug now, debugging is almost immediately possible.
My small organization is noticing output increasing. We're excited about it. I’m not sure about 10x… Like others have mentioned, it’s difficult because you have to measure different workloads.
I build things I never would have. My tooling is better and more robust than ever. I verify and test my work better than ever. I fix more bugs than I used to simply because no one needs to care if it fits into a cycle. I explore and solve more problems in more parts of the application, even if I don’t write code. I take better care of our infrastructure. Performance goes up, bugs go down, AWS resources scale back, costs go down. I’ve paid for my AI usage in scaled back resources several times over at this point.
It might not be 10x but it’s a significant multiple.
It is frightening to see how the latest anti-VPN developments in Russia [1] are somewhat replicated by some US states. For those not following, VPNs are widespread in Russia after the government started blocking popular messengers such as Telegram and websites such as YouTube, and a few weeks ago Russian government instructed major Russian websites (banks, VK social network, etc.) to stop serving users if the website detects the user uses VPN. I read this Utah news as something very similar: an attempt to reduce VPN usage by forcing the websites to collaborate. Feels very wrong to me.
Not trying to defend App Store policies, but writing this just for those who are struggling with Guideline 4.2 trying to publish an app that is only intended for a small group of users. There is a less well-known option called "unlisted app distribution", similar to unlisted YouTube videos: the app is public and can be downloaded using the direct link, but it cannot be found in App Store search. The "small, or niche, set of users" guideline normally does not apply for such apps.
To request unlisted distribution for your app, send it for review as usual, then file a special form [1], and mention that in the review notes.
Source: I struggled with Guideline 4.2 when I tried to publish an app showing the bell schedule and other local information for the neighborhood school. Its audience is, indeed, not of Apple scale: the school parents living nearby. Apple refused it as 4.2 and only agreed to publish it as unlisted, which I was okay with, because sharing the link between the parents was not a big deal. Google had no problems with publishing the Android app normally though.
> I struggled with Guideline 4.2 when I tried to publish an app showing the bell schedule and other local information for the neighborhood school.
Why would you not just make this a webpage, and then the users could add it to home page as if it were an app? no Apple review necessary then. What does it being an app give you besides bureaucratic headaches?
> 2. Those people and many more besides have no idea what "add it to home page" even means.
If Apple supported the beforeinstallprompt event (available in Chrome since 2015) then people would have same experience as installing app [0]. Instead, you must create a wrapper around webpage and submit thru App Store.
The main driver for making it into an app and not just a web page was the need to send push notifications. Of course, I just needed it for myself: hey, it's time to stop working and start driving to school to pick up the kid – "notify me 30 minutes before the last period ends" given that the schedule is different every day; then I just shared it with other parents.
There is a web version (it's Flutter so it was easy to make one), but parents use the app much more often.
Look how they hid the option to add a PWA to the home screen.
Instead of placing the option next to "Add to bookmarks" after you press "..." (especially if it recognized the page as a PWA!), you have to choose "Share" (which makes no sense!) and then scroll up (at the same time it's non-obvious that you can even scroll up) past the option "add to bookmarks" (again) and then pick the 5th option "add to home screen".
there still a lot of jank. on ios u can only doo this with safari, and even then u loose actual safari niceties like trad browser ui. and idk why but it op ens link in actual safari even if its the same app. unless u have a single page app that does nothing this is not a viable route/.
I live in the EU and just wrote my first PWA and that‘s not true, there is (almost) NO browser ui/ux.
No url bar, no back/forward, no tabs, nor translation, no menu bar, no loading indicator, just… pressing down on a link shows the target url and offers open, copy link, add to reading list and share -which honestly looks like a weird oversight.
I really hate it when the browser UI gets hidden. It loses me several great features like the ability to open multiple tabs. Or have a bookmark directly to a subpage.
Unfortunately some other features are only available to PWA do it's a tradeoff either way.
I also have a small private app that technically could have been a PWA.
It’s not a PWA because the UX is just always inferior. Even though we’ve come really far in browser UIs, the browser is still very clunky compared to the smoothness of a native app.
I assume you would use Oracle Cloud if, for whatever reason possibly related to legal or competition, you cannot use AWS, or GCP, or Azure. It's hard for me to imagine a startup that needs cloud and would onboard to Oracle Cloud and not to any of the top 3 providers instead.
I actually like the Claude's Co-Authored-By: line very much. Even in my personal repositories, where I'm the sole author and the sole reader, I would like to know if my older commit I'm looking at was vibe coded, implying possibly lower quality or weird logical issues with the code.
So, my personal rule is: if I implemented a feature with Claude, I'll ask it to commit the code and it will add Co-Authored-By. If I made the change manually, I'll commit it myself.
One thing I learned over the years is that the closer my setup is to the default one, the better. I tried switching to the latest and greatest replacements, such as ack or ripgrep for grep, or httpie for curl, just to always return to the default options. Often, the return was caused by a frustration of not having the new tools installed on the random server I sshed to. It's probably just me being unable to persevere in keeping my environment customized, and I'm happy to see these alternative tools evolve and work for other people.
This sort of thing is a constant tension and it's highly likely to be a different optimum for every individual, but it's also important not to ignore genuine improvements for the sake of comfort/familiarity.
I suspect, in general, age has a fair amount to do with it (I certainly notice it in myself) but either way I think it's worth evaluating new things every so often.
Something like rg in specific can be really tricky to evaluate because it does basically the same thing as the builtin grep, but sometimes just being faster crosses a threshold where you can use it in ways you couldn't previously.
E.g. some kind of find as you type system, if it took 1s per letter it would be genuinely unusuable but 50ms might take it over the edge so now it's an option. Stuff like that.
reply