マッチャ (Matcha)
@matcha-p.aoitori.mobi
And this is why trying to treat these things as "technology" is the wrong approach.
I've had way more success by treating them like an entry level call center worker. Provide documentation, interaction guidelines, scripts, etc.
You can build incredibly powerful and reliable tools this way.
I've had way more success by treating them like an entry level call center worker. Provide documentation, interaction guidelines, scripts, etc.
You can build incredibly powerful and reliable tools this way.
And that's the the crux of the problem. When my computer crashes, it stops working. When there's a render error, I can see it. Hell, if my heat stops working - my heat stopped working.
LLMs appear to "work" all the time, and the failure mode is only experienced later as real-life consequences.
LLMs appear to "work" all the time, and the failure mode is only experienced later as real-life consequences.
December 16, 2024 at 3:23 AM
And this is why trying to treat these things as "technology" is the wrong approach.
I've had way more success by treating them like an entry level call center worker. Provide documentation, interaction guidelines, scripts, etc.
You can build incredibly powerful and reliable tools this way.
I've had way more success by treating them like an entry level call center worker. Provide documentation, interaction guidelines, scripts, etc.
You can build incredibly powerful and reliable tools this way.
It's so satisfying logging into @bsky.app and seeing that the most recent post in my feed was 5 hours ago.
Nothing being shoved down my throat. No forced engagement. It's excellent.
Nothing being shoved down my throat. No forced engagement. It's excellent.
December 1, 2024 at 10:49 PM
It's so satisfying logging into @bsky.app and seeing that the most recent post in my feed was 5 hours ago.
Nothing being shoved down my throat. No forced engagement. It's excellent.
Nothing being shoved down my throat. No forced engagement. It's excellent.
Reposted by マッチャ (Matcha)
