Careful though: a lot of people are getting the feeling of getting 20x done. Do you have objective measurements?

In my case is nearer to ∞x. I have developed an opensource Android app which has already ~200 users that I would never ever written in my whole life. Zero experience with mobile development and zero time in my free time to focus on this appropriately to at least try to learn how to do it. I know myself, I would have given up before getting the first dummy APK on my phone. And while it's totally vibe-coded, in the sense that I just prompted CC and not written a single line of Kotlin code, I put a ton of effort on it anyway, on how I want it to behave, how it looks like, squashing all the usual subtle bugs that CC leaves here and there.

Did you prompt it step by step or let it do its thing for a bit?

I've been working on it since December. The first working prototype was a oneshot of a biggish enough prompt (for Opus 4.5). After that well, I didn't save the prompts (my bad) but I have probably prompted what... 2-3k 80-columns lines of English in Claude Code? Yeah, I guess we are in that ballpark. Sometimes it nails the new feature at first attempt, sometimes it takes a few attempts and corrections (and in that case it can definitely be frustrating)

What does it do?

(I'm going to break the rule I had to keep my HN identity separated from my real one with this post but here it goes...)

It's a frontend to the vehicle data served by TeslaMate (a local logger for your EV data), as a more mobile-friendly alternative to Grafana.

How do you even begin to define objective measurements of software engineering productivity? You could use DORA metrics [1] which are about how effectively software is delivered. Or you could use the SPACE Framework [2] which is more about the developer experience.

1. https://cloud.google.com/blog/products/devops-sre/using-the-...

2. https://space-framework.com/

I don't have time for that mysticism. I just know.

IDK just yesterday I got a complete slide / powerpoint-lite editor in Qt Quick that is sufficient for the use case I have in two prompts, roughly 7 minutes. How long would it take you to write, on your best day, using your favourite programming language ?

Ha ha, for some of us the "feeling" is good enough.

this is the most self-aware, honest answer IMO

For me the evidence is I have completed side projects I never would have before. I also recently started building a game that I had put off for years. At work I am closing more features than historically and at the end of the day not as fatigued. It’s only my experience and everyone’s is going to be different.

>Instead of engineers inventing 50 different frameworks and conventions for any given language or platform, maybe that energy will be directed to creating better AI tools.

I think this will happen since one of the reason for new frameworks and languages was improving the human experience of coding, but now that friction goes away and AI doesn't feel that.

Although we might need to study which language AI is best at, and possibly invent new ones to maximize that.

> Claude Code has killed my ADHD and turned me into an always-on hyper-focused machine.

> I am getting 20x done. This is a literal superpower.

Adding this comment to favourites to revisit in half a decade.

I've already "made fun" of your exaggerated hype comments, so I'll use this opportunity to say that I hope you remain sane and grounded in your discoveries. You wouldn't be the first to go psychotic after interacting with these stochastic parrots.

Don't you have anything better to do with your time?

I told you people back in 2019 that these models would replace Hollywood and you and others have been calling me all kinds of names, and every step of the way calling me an idiot. I'm a filmmaker - I know what I'm talking about. And now we're almost here. We have million dollar VFX services at our disposal for pennies.

Claude Code is doing the exact same thing for software engineering. I've been a senior software engineer for a good while - these capabilities are otherworldly and they can generalize to all new unseen problems. You're not paying attention.

I'd be more worried about whether or not you have a job in 5 years than whether I have or have not created a business or whatever criteria you want to use to thumb your nose at me.

You know how you can quickly ideate software plans for some large scale idea? Architecture, infrastructure, data models, etc., but the implementation takes longer? Claude Code short circuits that last bit. You need to hold your nose so you stop smelling whatever you're smelling and just try the damn tool.

I wish I could slap sense into you grumpy folks. You're so stiff in your beliefs. This is a train headed your way. Pay attention.

> I told you people back in 2019 that these models would replace Hollywood

What kind of alternate reality are you living in?

I wish you would disclose your credentials (though I admit privacy is an inalienable right of yours) so I could place the biggest AI hype-man on this forum. Actually, there is hype, and there is being completely gone with hubris and you’re towards the latter end of the spectrum, given your doomsday calls on other comments that software engineering is done for and that you believe AI is close to ‘putting all the HN engineers out of work’ (https://news.ycombinator.com/item?id=47185284)

> I wish I could slap sense into you grumpy folks. You're so stiff in your beliefs. This is a train headed your way. Pay attention.

Lay off the violent thoughts and get some rest, man. Sounds like you need it.

part of my adhd/bipolar is reacting like the guy you're replying to and I was thinking the same. Comment reminded me of when I'm in the "YES THIS IS IT" mode which usually isn't far off from hitting the wall. Hopefully just projection on my part, though, and this guy is really doing well. When I start talking like him though I usually have to take a step back and it'll be a topic in therapy next session.

100% feel the same way, and had the same starting point lol