make
I build mobile applications that feel quietly considered — focused on state architecture, API integration, and the small performance details users never see, but always feel.
Most of my recent work sits at the intersection of Flutter and applied ML — group-photo attendance, GPS-based sanitation tracking, ML-routed legal intake.
Currently learning more about platform channels and BLE audio at Bose Professional. When I'm not coding, you'll find me sketching architecture diagrams in notebooks or chasing a quieter way to ship.TrackIn detects faces in a single group photo, matches them against the enrolled roster, and writes attendance to Firestore — replacing roll-call with a snapshot.
Open to freelance, collaborations, and full-time roles. Most replies happen within a day.
aniudupa15@gmail.com→