←back to thread

9 points thebordella | 2 comments | | HN request time: 0.425s | source

As an Old, I came up learning the OG web dev languages - HTML, CSS, JS, PHP, etc. As a solo dev, I still largely work with these as natively as possible. But I am not a total dinosaur. I embraced jQuery in the early 2000's, Bootstrap scaffolding, and now often build reactive UI's using Vue 3.

Still, I continue to be picky about the many abstractions-of-abstractions that litter the dev landscape, mostly devised to drive their founders' business models.

AI-driven co-pilots are red hot now, and I want to evaluate their usefulness to my style of dev work. I've begun dabbling with services like Codeium and Claude, particularly within VSCode.

At first these AI tools seem exciting. A whole function conjured out of thin air from a short prompt! But the more I test drive, the more I find that I'm spending my time trying to craft prompts that are accurate enough to generate the code I want. Whereas I could be using that time to just conceive and write the code myself.

If you are not already well-versed in your coding languages, these AI tools must save a ton of time and effort. My question is - what about you devs who are already proficient in your languages of choice? How do you leverage AI co-pilots to maximize their usefulness without just falling into the "distracted by another abstraction" hole?

1. abelanger ◴[] No.42168963[source]
I've found that I rely most heavily on LLMs when:

1. I'm developing a utility package that's easily testable and I'm certain of the interface. I'll write the interface for the package in my editor, then ask an LLM to generate unit tests. Then, I'll sketch out the function calls/structure of the package and get an LLM to fill out the rest.

2. I'm bug bashing and want to quickly check if the bug is obvious. I'll feed a description of the behavior into GPT/Claude along with the relevant code (generally as little code as possible to prevent listicle-type responses that are irrelevant).

3. I'm adding code that follows an established pattern within the codebase -- for example, adding a new API handler that involves generating an OpenAPI path + component snippet, an http handler, and a few database methods. This is when copilots are particularly useful.

4. I'd like a sanity check for issues in a complex bit of code I've written.

I find these mirror the tasks you'd typically hand off to a less experienced dev on the team -- these are things that require validation based on knowledge that you already have, and the validation is more efficient than doing it yourself.

replies(1): >>42181670 #
2. hehehheh ◴[] No.42181670[source]
A new repo "find me the code that..." is an excellent use case