I build systems that see, listen,
and decide.
I'm Vighnesh Ganti. I like turning messy real-world signals — MIDI files, camera frames, circuit boards — into models that make a useful call. Below are two projects that ship end to end: from raw data to a trained model to something you can actually click or watch move.
Selected work
02 / projectsPiano Difficulty from MIDI
A transformer-based classifier that predicts how hard a piano piece is to play — from the MIDI alone. Uses CORN rank-conditional heads for ordinal regression across 11 difficulty grades, reaching Test MAE 1.131 on the PSyllabus benchmark.
Automated PCB Component Detection & Pick-and-Place
YOLO11 instance segmentation finds 20 classes of electronic components on a circuit board, maps pixels to printer coordinates, and physically picks and sorts them with a vacuum gripper on a modified 3D printer. Synopsys Science Fair 2026.