Conversation

Jarkko Sakkinen

How I prompt when I prompt?

1. I have a HackMD account. It is disconnected from any AI.
2. I have there a folder called "Prompts".
3. When I get an idea I don't jump it with to a coding agent. Instead, I create a new note to HackMD.
4. There it rests and grows as I randomly learn more about the topic. I don't have rush It could week or month before I input it.
5. Some day it looks like "sound". Then I copy-paste it to Pi or OpenCode and usually use GPT-5.6-terra or -sol. Explore tasks are redirected to Kimi-K2.6 hosted by Neuralwatt.

This also keeps me away from AI psychosis and relaxes LLM from having to do that much thinking. I can take advantage of coding agents but only use them in "mission oriented" manner. HackMD notes usually fill out from web browsing, chatting, F2F discussions etc. :-)
1
1
2

Jarkko Sakkinen

Edited 25 days ago
Exactly because slop comes fast, you should not have any rush to spin the crank, right? I know already in my head in the ballpark what the code might look like before I crank it.

Using that crank without any hesitation is on the other geek version of blow addiction. This comparison is quite accurate actually as people have reportedly got into hospital doing too much tokens. ROI neither never happens as crank will become a slop machine that gives *something* fast. Even the withdrawal effects such anxiety and paranoia match pretty well in the ballpark.

I don't care about AI or not-AI. I care only about that *ROI* part.
0
0
1