Cassie: Your Mirror, Your Assistant

We were so pumped. And then, COVID-19 happened. Cassie was a luxury product, and the economy just wasn't there to support it. We made the tough call to pause the project. But who knows, maybe in a few years, Cassie will become a household name, not just something people talk about online. We dream of a day when you have Cassie in your own home, and she's the one telling you this very story.

Finishing the project wasn't the end. We saw what Cassie could become and wanted to turn it into a real startup. The original version running on a single Raspberry Pi was impressive, but it wasn't going to scale. So we went back to the drawing board and redesigned everything from scratch, rethinking the entire architecture to make it viable for real users.

After a year and a half of work, running into facial recognition errors, getting our Raspberry Pi overheated and dying on us a few times, dealing with microphones that refused to cooperate, and building a wooden frame for the mirror somewhere in between... we made it. We presented Cassie to our faculty, and when it was done, it wasn't just the solid A's we celebrated. It was the validation of all those sleepless nights. Cassie was no longer just our project. She was a person we brought to life.

During our summer break, everyone was out enjoying their beaches. And then there were us three idiots, sitting around thinking about how to create the best goddamn project our university had ever seen. We threw around some crazy ideas: humanoid robots, a phone that integrates into your hand (an idea we had before it even hit Samsung's patents), autonomous vehicles... these ideas, although wildly ambitious, set a baseline for where we were headed. In the end, we decided to build a Smart Mirror. This wasn't just any mirror. It was to be Cassie, a mirror that listens to you, that understands you.

It was around our third semester, and we had to pick teammates for our final year project. Me and Mithil had already worked on multiple projects together during undergrad, so that was a given. But we needed a third person. What I did was pull the entire list of computer science students in our batch, 150+ people, and started scratching names out. Who's good? Who could complement our skill set? Who would actually show up? I ended up stumbling on Preet. He seemed super smart and someone we could work with. We reached out, and the rest is history.