alt Hacker News> It makes me less eager to create which is quite unfortunate.
It's been the exact opposite for me. Coding assistance is a great boon towards productivity simply because otherwise I wouldn't work on any of my old ideas stashed in numerous note taking apps. It's way easier today to go from 0 to something like an MVP, and see if there's something there. If there isn't, not much is lost. But without these tools it would be 0 all across the board.
Replies
I feel the same way. Things used to take so long just to do all the boilerplate. I would frequently get overwhelmed by the fact that I had to write each of the uninteresting parts to get to the interesting parts.
I've recently started using a Chrome extension of my own making[0] that allows me to block and highlight users on Hacker News. I remember trying to do this once a long time ago and it was so much work. I had to learn about the Chrome manifest's possible permissions and I had to format my options page nicely with CSS and I had to learn how to make the extension connect to a web page.
Same with these tools I've built for our family. My wife wanted to be able to give an AI some rough notes and have it polish things up using the notes she already had. She wanted to use ChatGPT. I know the theory about the thing:
* ChatGPT has custom GPTs
* Custom GPTs have Actions that call APIs
* If you have an OpenAPI spec, custom GPTs can understand how to call them
* If you have the HTTPS server, custom GPTs can access the endpoints
Previously, I'd slog through each step one by one. This time, on a day when I was watching my infant daughter, I managed to finish the whole thing fully functioning during a short period that she slept! Eufy Baby iOS app up in the corner of the screen, claude code on the left, ChatGPT in Chrome on the right. Knocked it out in an hour and my wife uses it every day.
Astounding tool. Then yesterday I wanted to print an alternative mount for our baby monitor so I can place it in a specific place. I couldn't find the camera mount STL files anywhere. 15 minutes for my wife to find my calipers, 5 minutes measuring, then untold GPU hours but zero of my time as codex built me a test mount[2] while verifying it against a mesh-contiguity script.
And my wife is a graphic designer, so she iterated on our wedding clothes by using Dall-E to design the ideas she had, polished it up in Adobe CS, and then we got an embroiderer we know to put it on my sherwani[3]. At first I thought that perhaps it was just us engineers who get a lot of value out of these tools, but my wife uses them to design things to make on the Cricut, or to help with stuff to 3d-print on our Bambu, and both of us have used it to come up with modifications to recipes which have had surprisingly decent effect!
Dude, my life is like 10x better, maybe 100x better. Everything I dreamed of is no longer gated by my lack of specific skill. I am only gated out by my taste.
1: https://wiki.roshangeorge.dev/w/Blog/2025-10-17/Custom_GPTs
2: https://wiki.roshangeorge.dev/w/Blog/2025-12-01/Grounding_Yo...
We watch a ton of stupid Hallmark Christmas movies during December, it's our thing. Very few of them are available here and don't have subtitles in our language, english subs do exist.
So, how hard can it be to translate them automatically? Tried a few ready-made tools and they either just plain didn't work or made stupid mistakes due to bad prompting.
I fired up Claude and had an MVP that could 1) rip english subtitles out of a mkv file 2) shove them to OpenAI API for translation in batches within a few hours
v2 took two evenings of me watching TV and bouncing around between Claude+Codex+Crush(GLM-4.6)
1) Get subtitles from original
2) Feed full subtitle file to GPT-4o for first pass analysis. It finds names for people, locations etc and decides a singular translation so they don't vary and gives a general context for the tone/genre of the movie for translation
3) Give gpt-4o-mini the first pass context file and a section of the subtitles in a loop
4) save them as .srt
The results are SO MUCH better than most of the "real" translation we've encountered. I think a bunch of them were done by a zero-shot AI or a human who didn't give a fuck about quality and had to meet a quota.
And none of this would've been done without AI, there's so much crappy boilerplate on top of the few actually interesting bits I would've just tolerated the bad ready-made solutions instead.