Why I Built SAGE

Over the last few months, I've been building an AI-powered TV show recommendation project.

During that process, I noticed something interesting.

The AI wasn't just recommending shows. It was doing a surprisingly good job of connecting the shows I like with my personality traits, business philosophy, relationship philosophy, and the way I tend to think about life.

I found that more and more interesting as time went on.

The 2 a.m. test

For whatever reason, that kind of self-discovery and psychology-adjacent stuff has been sitting at the front of my brain lately. I'd already been thinking about building something around that idea, but then Instagram did what Instagram does.

At about 2 a.m., I stumbled across an archetype test.

The ad caught my attention, so I clicked it. I went through the whole thing and answered what felt like 50 questions. I honestly thought that at the end I was going to get some kind of interesting discovery about how my brain works, how I think, what drives me, or something along those lines.

Instead, I got a $29.99 paywall.

The results were locked.

I didn't care that much, so I left.

But I was still curious.

So I took a screenshot of the ad that got my attention in the first place, dropped it into AI, and asked if it could create a test that would isolate those archetypes and figure out what was going on underneath the surface.

It did.

Then I took the test myself.

And honestly, it was surprisingly accurate. Maybe even more than accurate, it was revealing. It made me think about myself in a way that I found genuinely useful.

The missing piece

That was the missing piece.

A personality test, archetype test, or self-discovery report might be interesting, but if it just gives you a label and stops there, the value is limited.

So I did what I apparently do all the time.

I crawled my ass out of bed, went to the office, and started building a Custom GPT that could recreate that kind of discovery conversation for other people.

But once I got into it, the idea changed.

It wasn't enough for SAGE to just tell someone what it discovered.

It needed to give them something useful at the end.

Why it works the way it works

I know the process is a little clumsy.

You use the SAGE GPT. Then when it finishes, you copy the handoff and prompt it gives you. Then you save it somewhere on your phone or computer. Then you paste the Daily Coach Prompt into a new ChatGPT conversation when you want to start using it.

I get it.

That's not as smooth as a polished app with accounts, dashboards, saved profiles, and automatic memory.

But here's the reason:

It's free. Truly free.

I'm not collecting your data.

I'm not storing your answers.

I'm not charging you $29.99 at the end.

I could build this into a real platform and make the whole thing easier, but that would cost money every time someone used it. I can't offer something for free if it costs me significant money to operate.

So yes, it's a little clumsy.

But it's free.

And honestly, it's not that hard. If saving prompts and handoff documents isn't something you've done before, it might feel annoying the first time. But everything feels hard until you do it once, or twice, or twelve times.

That's the only feedback I get

There's no dashboard on my end. No collected results. No hidden analytics about your answers. Nothing.

If it works for you, I only know because you tell me.

If it doesn't work for you, I only know because you tell me.

If you loved it, hated it, felt like it nailed you, felt like it missed completely, or thought one part was confusing, come back and say that.

I'll use that feedback to make SAGE better.

Right now, I think it's solid.

I don't think it's perfect.

But I also don't know what needs to change until real people use it and tell me.

That's why I built it.

Not to sell you anything.

Not to collect your information.

Not to build some giant platform.

I built it because I had an experience that made me think, "This could actually help some people."

So I made it free and put it out there.