71–80 of 93 posts
Earlier quoted context omitted.
Biking isn’t difficult, but you’ll still travel more with a car.
Now imagine if your car didn’t have a steering wheel or pedals, but instead you had to explain to it where to go and it would misunderstand you 60% of the time.
Claude has helped me a lot to get through my small feature backlog faster. But if you don't review the code you're asking for hurt. It has literally written stuff like "if pending changes > 20, delete all", etc.
If left on auto pilot, it will sometimes get frustrated with bugs and start "cheating" in various ways (e.g. unsafe type casting, commenting out code/tests, etc.).
Earlier quoted context omitted.
IDK, there are a lot of little chores that I find needing done that previously would have been ten minutes of research and cli fiddling that are a prompt and copy paste now. It doesn't feel that different but it is a little faster
If you don't understand what the copy/paste is actually doing, you shouldn't be script-kiddying it in your project. If you do understand, then the AI is just quickly writing what you already intended. If you are having to understand something you didn't understand, it's probably taking a bunch of time to read and verify what the script does. This can be a good learning experience and reveal unknown unknowns, but prob…
Earlier quoted context omitted.
No, there's a very very long tail of tasks that are too niche for a proper tool to exist for but it would be annoying effort to figure out yourself. My most recent example of this: write a script to delete all git branches that point to HEAD. Go ahead see how long it takes you to figure out the command for that, because it took me under 60 seconds with AI.
This is the learning to fish problem. If you understand git well, a solution isn't very hard (eg, rev-parse each branch and compare to the rev-parse of the current branch and deleting if they match). If you don't understand git well, you shouldn't be turning into a script-kiddy pasting in whatever the AI spits out and hoping that it works.
Learning about `git rev-parse` through documentation and learning about `git rev-parse` through AI fundamentally have the same outcome at the end of the day: you have learned how to use `git rev-parse`.
Earlier quoted context omitted.
The one time so far that I actually implemented a full feature using only vibe coding (and it wasn't trivial), I had to be very specific about what I wanted, and it was in a very widely used stack (React/NextJS), it was a very common feature (sorting, filtering, pagination on a table) and it worked beautifully. But it felt magical the one time it happened. Every other time, I'm mostly using tab autocomplete and that…
Your last sentence is a bit of a gut punch tbh - I was meaning to get more familiar with Elixir for side projects, and I’ve made a few small web apps with LiveViews. Absolutely love it, even though I don’t feel I‘ve grasped the language fully yet. But some research shows that most LLMs don’t support Elixir that well. But since my free time is so extremely limited, it would be really nice to get better LLM and agent s…
I have already vibe-coded countless number of scripts and two side projects that were on my "list" for years, these projects would have never seen the light of day if not for Cursor. That's the whole point.
The fact that they could be "vibe coded" means that they (or something similar) probably already existed, but you just didn't know they did.
Earlier quoted context omitted.
This is the learning to fish problem. If you understand git well, a solution isn't very hard (eg, rev-parse each branch and compare to the rev-parse of the current branch and deleting if they match). If you don't understand git well, you shouldn't be turning into a script-kiddy pasting in whatever the AI spits out and hoping that it works.
You are using "script-kiddy" as an ad hominem rather than making an argument with proper justification. "script-kiddy" refers to website vandalism and cracking. That has nothing to do with using AI to learn what rev-parse does. Learning about `git rev-parse` through documentation and learning about `git rev-parse` through AI fundamentally have the same outcome at the end of the day: you have learned how to use `git r…
But with a non-zero chance of hallucination.
I’ve been able to extend its functionality several times without it breaking, and had Claude separate all functionality as components so it’s easier to manage. I could never have done that myself.
I consider it a very good prototype although we use the tool internally on a weekly basis. If we’re going to continue using it long term, I will probably find an experienced programmer to rebuild it from scratch, and support it.
Earlier quoted context omitted.
Your last sentence is a bit of a gut punch tbh - I was meaning to get more familiar with Elixir for side projects, and I’ve made a few small web apps with LiveViews. Absolutely love it, even though I don’t feel I‘ve grasped the language fully yet. But some research shows that most LLMs don’t support Elixir that well. But since my free time is so extremely limited, it would be really nice to get better LLM and agent s…
Unfortunately this is consistent for non-ultra-mainstream langs. They’re next to unusable for complex Haskell. If you want to vibe code, it’s Python/typescript/react or nothing.
This is not news for people who already know how to code. The problem is that there are tens of thousands of people who have no idea what they're doing, yet they evangelize their hollow victories to anyone willing to listen. Worse yet, there are grifters like levelsio who will pretend like it's great because they know that's what their followers want to hear.
This is a good take. The problem is not that vibe coding is bad or worthless, it’s just that vibe coding exposes new groups of people to new kinds of risks.