←back to thread

Focus on decisions, not tasks

(technicalwriting.dev)
293 points kaycebasques | 2 comments | 18 Oct 24 17:58 UTC | HN request time: 0s | source
Show context
simonw ◴[18 Oct 24 21:02 UTC] No.41883461[source]▶
This fits the way I like to use LLMs: I always ask them for options, then I decide myself which of those options makes the most sense.

Essentially I'm using them as weird magical documentation that can spit out (incomplete but still useful) available options to guide my decision making at any turn.

replies(2): >>41883780 #>>41886486 #
jiggawatts ◴[18 Oct 24 21:44 UTC] No.41883780[source]▶
I like to think of it as the apprentices working for famous artists like Leonardo. The master would draw the outline/sketch, and then the students would fill in the blanks under supervision. Sometimes, the master would steal ideas from the students.
replies(1): >>41886927 #
1. aldanor ◴[19 Oct 24 10:21 UTC] No.41886927[source]▶
Smells like reinforcement learning in real life. Master sets ups the task environment, collects samples from students, picks the best and maybe even augments them. Students watch the master and learn... and the cycle continues.

And then the master becomes a grandmaster (unless entropy explosion occurs).

replies(1): >>41888673 #
2. gtirloni ◴[19 Oct 24 16:16 UTC] No.41888673[source]▶
Yea, and it's what ChatGPT does sometimes even if you don't ask it for multiple options. It shows two options and asks you to select the best one. Essentially using customers to help with training.