Thinking out loud, and what I learned
A journal of what I built, what failed, what surprised me, and what I'd do differently. This is where I work through ideas in writing — not to present finished conclusions, but to think more clearly.
Why My First Autonomous Vehicle Failed
My initial model of how sensors worked was completely wrong. The errors weren't random — they were predictable, once I understood what I'd assumed too quickly.
What I Learned Building a Fall Detection System
Designing a reliable classifier for human motion — on a microcontroller, in real time — taught me more about the sensitivity/specificity tradeoff than any textbook could.
Can AI Actually Help Parents of Autistic Children?
When I started building a resource platform for parents of autistic children, I assumed the problem was technical. It turned out to be mostly about language, trust, and mental models.
What I Learned From My First Computer Vision Experiment
Computer vision tutorials make it look easy. Reality has different lighting conditions, different backgrounds, and hardware that isn't quite fast enough for what you're trying to do.
A Problem I Set Out to Solve
There's a navigation problem I've been thinking about for months. I explored several approaches, identified their limitations, and worked through the challenges until I found a way forward. This is the story of the problem, the obstacles I encountered, and how I approached solving it.
What I Would Build Differently the Second Time
After two major projects, there are patterns of mistakes I keep making. Writing about them is the best way I know to stop repeating them.