By JT White, VP Content Studio
I spent last week at an AI Residency with our product and AI teams, and honestly? My brain is still buzzing. We spent the better part of five days deep-diving into the how and when of leveraging AI. It was one of those rare, wonderful weeks where you step back from the daily grind to look at the horizon.
As I’ve been processing everything we built, broke, and discussed, a few key themes kept rising to the top. Whether you’re a builder, a leader, or just trying to figure out where you fit in this "AI shift," here is what’s on my mind:
There are now a dozen different ways to get to the same endpoint with AI tooling. Whether it's prompt engineering, RAG, or fine-tuning, the path isn't linear anymore. For product teams, the potential to create workflows that enhance and streamline our actual thinking process is massive. We aren't just automating tasks; we’re augmenting the way we problem-solve.
One thing’s for sure: you still need humans in the loop. It was fascinating (and a bit humbling) to see just how wrong some of the AI’s assumptions could be. Without a subject matter expert to course-correct, you can move very fast in the wrong direction. LLMs have different strengths. Some are better at logic, others at creativity, but none of them replace the "gut feel" and context of a seasoned product person.
As a real life example, I spent most of the week messing with Claude Cowork. Being able to CODE something from prompts was not only amazing but wildly easier than even two months ago. But I did hit some walls. I ended up using Gemini Deep Think for a few tasks where it was just better. I leveraged some internally tooling our AI team has built to make the deployment process better. And I spent a lot of time getting better at Docker.
Each of these tools has its own flavor of interaction. None better than others, but all a bit different. I spent far more time fine-tuning prompts and consideration sets than I did actually messing with the actual code. And at the end, I had something meaningful. Production-ready? No. But a hell of a jumping-off point.
It has NEVER been a better time to be a builder. The distance from ideation to creation has vanished. But as a lifelong product person, I’ll be the first to say: that might also be a terrible thing.
Just because you can build something in an afternoon doesn't mean you should. The friction we used to face in development often acted as a filter for bad ideas. Without that friction, the burden of thoughtfulness falls entirely on us. Take the time. Be intentional. Don’t let the ease of creation replace the discipline of strategy.
This is an exciting time to be alive, but it’s also a weird, transitional moment. If you are leading an organization:
If I walked away with one mantra, it’s this: be more curious than convinced. The people who think they have AI figured out are the ones who’ll be left behind when the next model drops. Stay open, keep experimenting, and don't be afraid to admit when the tech gets it wrong. It is an accelerant for sure. But it’s not a replacement for critical thinking. You still need to DO THE WORK. And you should want to, because if leveraged intelligently I think you can get more good things done faster than ever.
It’s a wild time to be building. Let’s make sure we’re building things that actually matter.