Ask HN: How do you as a normal swe use AI? | Hacker News Reader

Ask HN: How do you as a normal swe use AI?

5 pointsatworkc9 comments

So am looking for more normal people, experienced people using AI, but I feel like whenever I hit youtube or other platforms the AI culture is way deep in the trenches and don't feel like this is how the majority of people tackle work, especially at their jobs.

What I see is mostly solo-devs going full battlestar gallactica on things. And I just want to see other people, normal people talk about and show how they use these new tools.

I'm also more interested in the first principles of things and not necessarily the shiny new model or shiny new workflow or shiny new thing. More towards problem solving.

So, if you have a bit of time to walk me through how you usually work in a day or point me to some good videos, resources on regular people doing it I'll be grateful!

View on HN

I wanted to document this largely for myself (and because other folks have asked me this question), so I ended up writing a blog post on it: https://adamtaylor.me/posts/how-i-use-ai

I hope that doesn't run afoul of self-promotion rules. I just thought it'd be nice to have this answer somewhere it can evolve over time.

Definitely appreciate the write-up!

And I think I'm in a somewhat similar place with how I use the tools as yourself.

How do you feel about your workflow, do you feel you're underusing the tooling or overusing it?

Actually I feel like I'm in the goldilocks zone right now. I've tried automating more for the sake of automation, and didn't really feel any more productive. My workflows have grown naturally as I feel like AI can fill certain gaps, and as I watch other people use AI successfully.

It does feel like an organic process, like I have to intuitively feel the solution for the need I have, can't really use all the shiny things effectively if I don't feel the pain in my bones which they solve.

I've been moving my goldilocks zone a bit towards scheduled tasks simply because I've actually felt the need and pain they solve, more like I have some backburner things I want to take care of but they're so unimportant that I'm unwilling to spend more than 5m here and there thinking about them so I add a scheduled taks every other day which does some leg work, I review for 5 minutes and then forget about it until the next iteration.

I think you're 100% correct that you need to feel the pain that the agent is supposed to solve. It's like trying to explain to a junior the power of Rails if they've never even built a regular static HTML website. It's going to be very difficult for them to understand why things are the way they are.

I'm very allergic to "solutions just because we can". In fact, one way I use AI that may be slightly not-mainstream is that I use AI to actively pursue subtraction that I couldn't otherwise do: cutting down on all my tools I've built over the years, culling cruft: notes, scripts, comments, etc. until I'm left with a smaller footprint of stuff I know is useful.

AI as a replacement of Google. When I wanted to learn about something the usual path was: google it. Go through dozens of links, discard, filter the good ones, read them, and gain the knowledge.

Nowadays I ask the question to the LLM and for every answer I demand the links/urls where the LLM get the info from. I still read the referenced web sites. It’s faster.

Could you clarify why you would still go through the referenced website? I mean, let's say if you were to fix an Nginx file and you ran into an issue, would you still be going through the docs?

Just asking to understand if that's the best way to use AI, to keep our own brains working, or is there a different perspective on it?

I’ve seen where the LLM gets the right page, but jumbles of the facts and gives the wrong answer. If you want correct answers, go to the source material, not the AI summary. The source also tends to give additional information that provides context for a better mental model.

One could say it’s a way of “trust, but verify “

Reply on news.ycombinator.com