The better AI agents get, the less I want to chat with them.
Chat made AI accessible. But agents need a quieter interface: work in the background, bring the human back when judgment matters.
https://t.co/2RHlbiR6JL
Also known as tinkering.
It's the combination of expertise with play, and it's euphoric. This is why Claude Code and similar are so addictive (in a good way).
The distinction between vibe coding and hacking/building/etc. is just how much knowledge you have when playing.
LLM prompt tip:
If you want strict execution → give only the exact step.
“Run this command.” No extra context.
If you want the model to find workarounds → give more details about the problem.
Less context = follow instructions.
More context = problem solve.
Don't think of LLMs as entities but as simulators. For example, when exploring a topic, don't ask:
"What do you think about xyz"?
There is no "you". Next time try:
"What would be a good group of people to explore xyz? What would they say?"
The LLM can channel/simulate many perspectives but it hasn't "thought about" xyz for a while and over time and formed its own opinions in the way we're used to. If you force it via the use of "you", it will give you something by adopting a personality embedding vector implied by the statistics of its finetuning data and then simulate that. It's fine to do, but there is a lot less mystique to it than I find people naively attribute to "asking an AI".
@DisneyPlus Every time I try to watch something I get “Protected Content License Error” and need to sign out of my account and sign back in. If you want to grow subscribers, fix these technical problems.
@DonnyWals A little late, but better late than never 🤣 Thanks Donnie for keeping the community engaged. I’m learning Vapor and making a mock server. I ❤️ the idea of sharing code between the server and the client.
Swift engineering speak for “Have you tried rebooting it?” 🤣
1. Clean build folder
2. Delete derived data
3. Pod deintegrate/pod install
4. Update to latest SPM