The Pattern I Didn’t See
It started with figure skating.
I was watching the Olympics and this guy lands a quad. A quad. I didn’t even know they did those. I thought triple axels were still the ceiling. So naturally I go down the ChatGPT rabbit hole. History of the triple axel. When did quads become standard. Who was first. You know how it goes.
Then I asked the question that always gets me in trouble: “What are some examples of things that were considered impossible until someone did them?”
The four-minute mile. The sound barrier. Everest without oxygen. Powered flight. Heart transplants. Great list. Exactly what I expected.
What I did not expect was what came next.
Excuse me?
I didn’t ask for a personality assessment. I was asking about ice skating. And now this thing is telling me it’s been watching. Noticing. Connecting dots across months of conversations I’d mostly forgotten about.
Something about seeing your own scattered interests mapped into a pattern by a machine is deeply strange. Each of those conversations felt standalone to me. I wasn’t building a thesis. I was just… curious. But apparently my curiosity has a shape, and ChatGPT had been quietly tracing it.
I felt exposed. That’s the only word for it. Not violated, because everything it said came from things I’d freely typed into a chat window. But there’s a big difference between talking into what you think is a void and having that void suddenly describe your psychology back to you with uncomfortable accuracy.
It’s like walking past a mirror you didn’t know was there. You catch yourself from an angle you’ve never seen. You recognize the person, but something about having a machine describe you to yourself before you’ve described yourself to yourself is… a lot.
But it kept going. And I let it.
That last line landed hard. Not because it was some revelation, but because I recognized it instantly and wondered why I’d never said it that simply myself.
Then came the blind spots. And this is where it got really useful.
That last one stopped me cold. Because I’ve lived that exact dynamic dozens of times and always chalked it up to the other person being stubborn or not seeing the data clearly enough. It never occurred to me that the data wasn’t the problem.
Data Doesn’t Move People
Here’s the thing. That same morning, I’d been on a call with Blake, our Product Owner for Sentry. We were talking about how Sentry’s value is completely transparent. You can literally watch the recording and see the evidence yourself. No black box. No algorithm score you have to take on faith. The data is just… there. Right in front of you.
And the whole time I’m thinking: why doesn’t everyone just see this? Why do we need a light show? Can’t the evidence just speak for itself?
No. It can’t. Not for most people. That’s what ChatGPT was trying to tell me, and it’s what I’ve been slow to accept. Data doesn’t move people who are already emotionally committed to their current way of doing things. You need the spectacle first. The spectacle gets them in the room. The evidence does the work once they’re sitting down.
P.T. Barnum had something there. I used to laugh at that. I’m not laughing anymore.
Building Around Blind Spots
Ray Dalio writes about this in Principles, and it’s one of the ideas that sticks with me most. Your weaknesses are usually just the flip side of your strengths. You don’t fix them by becoming a different person. You fix them by surrounding yourself with people whose strengths are your blind spots.
I don’t do light shows. I’m never going to do light shows. And honestly, I shouldn’t try, because the minute I start faking enthusiasm for spectacle, I lose the thing I actually bring to the table, which is cutting through noise and building systems around what’s actually true.
But I need people who live for the show. People who get that emotion isn’t the opposite of logic. It’s how you get logic through the front door. The marketing, the storytelling, the conference presentations, the co-branded partnership rollouts, the case studies, the testimonials. All the stuff I tend to wave off as packaging? That’s not packaging. That’s the translation layer between what we’ve built and what the market needs to feel before it’ll believe what we’ve built.
Blake gets this. When he talks about wanting conference-level pushes with our clients, wanting case studies and co-marketing, that’s not him being a salesman. That’s him doing the part of the job I can’t do and honestly don’t even naturally see needs doing. And it’s not just Blake. It’s Meirah. It’s Theresa. It’s Yosef. The people on our marketing team who think in stories and emotion and spectacle are not doing something lesser than the engineering work. They’re doing the thing that makes the engineering work matter to anyone outside this building.
That morning, I told Blake he was the first person I was sharing this ChatGPT experience with. I told him my blind spot, the tendency to believe facts alone should win, was something the whole team needed to understand about me. Not as an excuse. As a design spec. Here’s where I’m strong. Here’s where I need you.
There’s something humbling about admitting your wiring has gaps. Nobody loves that. But there’s also something freeing about it, because once you name the gap, you stop trying to be everything. You start building something bigger than yourself.
The machine didn’t tell me anything I couldn’t have figured out on my own. But I hadn’t figured it out. That’s the whole point. Sometimes you need a mirror at an angle you’d never choose for yourself.
I know what I bring to CloudResearch. My job is to see through walls. I need people around me whose job is to make everyone else care about what’s on the other side.