←back to thread

A staff engineer's journey with Claude Code

(www.sanity.io)
548 points kmelve | 4 comments | 02 Sep 25 19:34 UTC | HN request time: 0.001s | source
Show context
furyofantares ◴[02 Sep 25 20:52 UTC] No.45108880[source]▶
I've come around on something like this. I start by putting a little effort into a prompt and into providing context, but not a ton - and see where Claude Code gets with it. It might even get what I asked for working in terms of features, but it's garbage code. This is a vibe session, not caring about the code at all, or hardly at all.

I notice what worked and what didn't, what was good and what was garbage -- and also how my own opinion of what should be done changed. I have Claude Code help me update the initial prompt, help me update what should have been in the initial context, maybe add some of the bits that looked good to the initial context as well, and then write it all to a file.

Then I revert everything else and start with a totally blank context, except that file. In this session I care about the code, I review it, I am vigilant to not let any slop through. I've been trying for the second session to be the one that's gonna work -- but I'm open to another round or two of this iteration.

replies(1): >>45109172 #
soperj ◴[02 Sep 25 21:17 UTC] No.45109172[source]▶
and do you find this takes longer or shorter than just doing it yourself from scratch?
replies(3): >>45109241 #>>45109277 #>>45109322 #
1. shinecantbeseen ◴[02 Sep 25 21:26 UTC] No.45109277[source]▶
I’m with you. Sometimes it really just feels like we’re just tacking on the cognitive load of managing the drunk senior in addition to the problem of hand instead of just dealing with the problem at hand.
replies(1): >>45109538 #
2. sfjailbird ◴[02 Sep 25 21:47 UTC] No.45109538[source]▶
A hundred times more time is spent reading a given piece of code, than it took writing it, in the lifetime of that program.

OK I made up the statistic, but the core idea is true, and it's something that is rarely considered in this debate. At least with code you wrote, you can probably recognize it later when you need to maintain it or just figure out what it does.

replies(1): >>45110580 #
3. adastra22 ◴[02 Sep 25 23:45 UTC] No.45110580[source]▶
Most code is never read, to be honest.
replies(1): >>45110779 #
4. furyofantares ◴[03 Sep 25 00:13 UTC] No.45110779{3}[source]▶
In the olden days I read the code I wrote probably 2-3 times while in the process of reading it, and then almost always once in full just before submitting it.