I am a teenager who would rather take something apart than read the manual. I believe engineering is not about getting it right the first time — it is about building publicly, failing visibly, and documenting every wrong turn so the next person does not have to start from zero. This site is that documentation.
A regenerative CO₂ preprocessor for lunar habitats. I started this in mid-March 2026 after an email from the STEM Innovation Team at Space Center Houston. Three months later I had a working prototype with live sensors and a web dashboard.
Read the build log →AI-driven robotic arm using Seeed Studio Grove Vision AI. C++ firmware with web-based servo control. On hold while I finish AERO-LITE, but coming back this summer.
Build log coming soonI write about what breaks, what fixes it, and what I wish someone told me before I started.
8 slides, and a prototype that tweets CO₂ data. What I learned about telling a technical story to judges who see 50 decks a day. Also: why you should never claim 90% efficiency if you have not measured it yet.
NASA Competition StorytellingThe SCD41 would not show up on the scanner. I checked wiring, pull-ups, addresses, and nearly gave up. Then I realized the multiplexer channel mask was wrong. A story about patience, Serial Monitor, and why sleep matters.
ESP32 I2C DebuggingI wanted to compress cabin air directly. Then I did the math and realized 99% of that energy compresses nitrogen, not CO₂. I felt stupid. Then I redesigned everything. That is where AERO-LITE actually became AERO-LITE.
Design Math Iteration