I've been a catcher most of my life, and calling pitches is one of my favorite parts of the game. These days I pitch in a men's league. Sometimes, I know what I want to throw (and I hate shaking off pitches). So I wanted to figure out how I could call my own pitches, like some MLB pitchers now do. I call that FIFO University: Figure It the Fuck Out.
Enter this little orange device.
I Had No Clue What I Was Doing
I didn't know where to start, so I asked AI to plan it and followed along. I ordered the parts I was told to order, put them together, and kept going step by step until it was time for the one part that actually mattered. Playing a call in my catcher's earbud.
It wouldn't connect.
Turns out the chip I'd been told to buy only had the low-energy kind of Bluetooth, and earbuds need the classic kind to play audio. The plan never checked that, and neither did I, but then again, that was the day I learned the difference between types of Bluetooth. Every step after it was built on a part that could never do the one thing the project needed.
I was pissed. Mostly at myself, because I had been nodding along to a plan I didn't understand, and when it broke I couldn't even tell you why.
So I threw out the AI's plan and made my own, based on how I thought it should work. One little device on my belt that holds my calls and sends them straight to the earbud. No middle steps I couldn't explain. That's the version that worked.
Build Week
Even the version I understood had its moments.
The first build played beeps instead of my recordings. I had four clips of my own voice sitting right there, ready to go, and the AI made up its own tones instead. Tones were never part of my plan, so my exact words that night were "always refer back to the plan." That was my first real lesson in building with AI. Quality check its work against the plan, every time. And expertise is leveraged and learned, not created out of thin air.
Then the audio would start and cut out. Once that got sorted, I was stoked. The AI got an honest status update as if it were one of my teammates: "dude, it's working."
The next day I had a game. It was too close of a game to mess with anything new.
But it got me thinking about how I'd actually use it during a game. Clip it to my belt, probably. Then the paranoia kicked in. With the screen facing out, what if the other team could read the pitch on it? (I know it's tiny and hard to read. I'm paranoid anyway.) So now the screen never shows the pitch. It just says "PITCH SENT."
300,000 Views Later
Honestly, it sat around after the first build. At my desk it was fine. Outside at about 30 feet, the call seemed delayed. The real test came on a family trip to Montana and Vermont, in a bullpen at the full 60 feet 6 inches. It was hit or miss. It still hasn't been used in a real game.
Months later, I posted a short video of it for two reasons. If a potential employer ever went looking for me, I wanted something real to show them. And for the coaches and players who don't have the money for a pro system, I wanted to show that you can figure something out for yourself.
Then the video took off. That is not normal for me.
People asked to buy one, asked for the parts, asked for "that doc." The funny part is that I got mine by asking AI how to build it, and now people were asking me instead.
I thought about selling it. Perfect chance to make some money, right? But it's hit or miss at full distance, and I didn't want to hand people a finished box anyway. I wanted to give them the chance to do what I did and build their own tools. I've never had the resources, and I still don't. I'm like a small market team, think the Rays, getting scrappy and creative instead of buying the best on the market like the Dodgers. This was a chance to help those in my shoes.
One guy in the replies had already built his own with a keypad so he could call location too. That's the kind of scrappy I'm talking about. He must have the ever elusive command I've been searching for since I was 9. Mine doesn't call location, because there's no need to call location if you don't know where it's gonna go. I aim for middle-middle and let my inevitable miss do the rest of the location.
I also have a lot of ideas that never get finished. When this one blew up, I took it as my sign. This is the one I finish. Well, this version of it anyway. Who knows, maybe there'll be an update someday.
So I've named it OpenSigns and I'm sharing it. Enough of the background story, here's the post about the actual build.
Enjoy.