Water Management

The Attempt

I need to be honest: this project was low-key haunting me. It wasn’t hard. That was the problem. It was obviously doable, clearly useful, and I just… didn’t. Every time I climbed stairs to check the tanks, a small voice whispered “you literally have the parts in a drawer.” Months of that. The problem: two water tanks, one alarm, zero information. The alarm philosophy is “something is full, figure it out yourself, also here’s some cardio.” The solution, obviously: microcontrollers, ultrasonic sensors, cloud dashboards, IoT everything. I am a software developer. I solve problems by making them more complicated first.

The Extraction

My younger brother visited. He’s not a programmer — he’s serving national duty, different world entirely. But he was curious. So I sat with him. Handed him the problem. Watched him struggle with syntax, with sensor logic, with the gap between what AI tools generate and what actually runs. I remembered what it felt like to not know yet. We got the core working. Motor control, safety cutoffs, the bones of something real. It’s still not installed. Needs electrical work. Needs a dashboard. Needs his next leave, probably. But the guilt is gone. It was never about the automation.