A browser-based platform for assessing and tracking cognitive and motor skills using everyday device sensors — camera, microphone, and touch.
"What happens when I use AI to turn everyday devices into tools that test, teach, and track human physical and cognitive function?"
20 sessions documented from first prototype to deployed platform — decisions, dead ends, and everything learned.
Architecture diagrams showing how Baseline evolved from a single-page prototype to a full-stack platform.
54 moments where AI suggestions were rejected or revised — and why those decisions shaped a better product.
Four technical decisions that compounded across the project — tracked across versions with annotated code.
Artist statement, process arc, the Five Questions, disclosure statement, and M6 completion checklist.
Sessions 1–2 · March 2026
First prototypes in Google AI Studio. Built VibeWord Animator, Dot Destroyer (bilateral hand tracking), and Reaction Time Test — discovered the quarter question through making.
Sessions 3–6 · March–April 2026
Expanded from single game to multi-test platform. Wrote Position Statement: accessible at-home assessment, unusual inputs, professional + approachable, NOT diagnostic. Moved to Replit.
Sessions 7–9 · April 2026
Built homepage and Bilateral Coordination Test. Established the dark-blue color system, card design, GAME_STANDARD.md. Solved the double-mirroring problem after 5 iterations.
Sessions 10–17 · April–May 2026
Supabase backend, authentication, data persistence, Chart.js graphs. Built Reaction Time Test, Rhythm Sync, Fine Motor Precision. User testing. Deployed to Vercel.
Sessions 18–19 · May 2026
Fixed RLS vulnerabilities, removed guest mode, patched XSS across 51 values. Renamed Finger Maze → Fine Motor Precision. Added Movement Control metric. Platform hardened and live.
Session 20 · May 2026
Created the complete process book documenting the entire 19-session arc. Compiled all Records of Resistance, system architecture maps, leverage points, and the M6 final artist statement and Five Questions.
Baseline is a browser-based platform for measuring human physical and cognitive function using the sensors already built into everyday devices — no specialized equipment required.
Three assessment tests are live: a Reaction Time Test using touch events, a Bilateral Coordination Test using camera-based hand tracking with MediaPipe Hands, and a Fine Motor Precision Test using camera-based maze navigation. Results are stored in Supabase and visualized with Chart.js, letting users track their performance over time.
This process book documents how Baseline was built across 20 sessions — from a vague idea about gestures and games to a security-hardened, deployed platform with a real backend. The documentation includes every major decision, dead end, and resistance moment along the way.