Embarrassing AI #60: Back on the Horse

Things built: 23 People watching: 19 Cashflow:-$72

I’m not gone, I was just resting.

Embarrassing, I know.

Also, I’ve been busy with a bunch of non AI related stuff so didn’t have that much to report. I never meant for this newsletter to be the source of busy work, rather a way to share interesting findings as I learn more about AI. I think you have enough spam. You don’t need any of mine.

In order to keep things feel light and fun, I’m changing my release schedule. Going forward I’ll have posts out on Tuesday, Thursday and Saturday. I tried batching work, but that doesn’t really work because I write these posts in response to things I discover or experience as they happen. Batching just turned me into a content factory. 🤮

When the spirit moves me, I’ll post more.

Especially when working on automations, there is a fine line when it is just that, automation, and where adding AI to the mix makes sense. Sometimes its just about creating workflows, with very specific mapping. It can be tempting to sprinkle in some AI, because it seems like the thing to do, but I find myself asking if that is necessary.

Do I gain something by adding an error prone source? Is an AI agent going to save me work by making decisions within the limits I can define or am I just adding chaos, assuming, hoping, praying that it will work out?

For me the answer is usually pretty straight forward. Am I spending more or less time making sure the AI is making the right decisions than I would need to make those decisions myself?

If I’m spending more time, often that means that I have not spent the right amount of time, or applied the right amount of rigor to define what I want it to do. Typically this means that I don’t know what I want to the point I can describe it.

That’s fine. That’s just being a human. We’re very good at fudging things, figuring them out on the go.

Many times, with AI, that’s also fine, because the decisions it makes are as good if not better than the ones we would, especially when it has context. What I mean is that there is a difference between not knowing what I want as an end result and not knowing exactly how to do it.

Without knowing a desired result, it’s difficult to get good ones out of AI. Sure enough, you can get lucky, but that doesn’t work out with any kind of consistency.

On the other hand, if you know what you want but are not sure how to get there, AI can feel like a magic solution.

I haven’t found a magic solution. Yet. Much of it is trial and error, which is part of what makes this exploration so exciting.

I’m working on it.

Tools I am using to fake being an expert.

1. Beehiiv (The Platform)

"I tried other platforms, but Beehiiv is the only one that makes growth feel like a game. It’s what runs this newsletter. The Deal: You get a free trial + 20% off (I get a commission to keep the lights on)."

2. Wispr Flow (The Time Saver)

"I hate typing. This tool turns my rambling voice notes into perfect text. It’s the only way I can write this much every day. The Deal: We both get a free month if you try it."

3. Emergent (The Coding Cheat)

"I am not a coder. Emergent is my 'vibe coding' partner that writes the messy parts for me. Essential if you want to build apps without a CS degree."

🧘‍♂ Last Word.

If what you’re doing feels like busy work, try stopping for a while and see what you miss doing.

Making daily progress,

P.S. Did I do something totally backwards today? Or do you have a better prompt for this? Hit reply and tell me. I’m here to learn just as much as you are. (I read every single email).