Hacking something impactful

I am building something in healthcare. And yes, I know how that sounds.

It is a heavily regulated domain, the stakes are unusually high, and building anything serious here is not a weekend project. But that is exactly why I chose it. I have been doing hackathons almost every month for the past year, and the pattern I keep coming back to is this: the ideas that are the most fun and fulfilling to build are the ones with the biggest potential impact, even when they do not win.

So when Google launched the Gemini Live Agent Challenge on Devpost last week, I knew I don't want to build another cool, but ultimately a toy project. I wanted to build something I could actually stand behind.

For me, healthcare is that domain. The potential to meaningfully change lives is greater here than almost anywhere else. And if I can solve the right problems and build a good solution, this could genuinely matter to people. That is the kind of project I am willing to burn the ships for. I know it's a risky bet, especially in the context of a hackathon. The prior Gemini 3 hackathon even discouraged it outright. But I think if the idea is strong enough, it speaks for itself.

So what is the idea? What am I planning to build and how?

I will be posting here every day, documenting the process of going from exploration, to raw ideation to something production-grade. Today is day one. The idea is locked in. Tomorrow, we get into what it actually is. Stay tuned for more.