TinyML in Action: Real-time Tomato Grading on ESP32

JeremyCMorgan · x · 2026-08-29

This project demonstrates an end-to-end TinyML workflow for industrial use. It processes optical sensor data to classify tomato ripeness, exports the trained model to embedded C code, and runs inference on an ESP32. Designed for industrial sorting lines, the system handles high-frequency spectral data from conveyor belts and triggers pneumatic ejection commands based on ripeness and size.

Original post →

More from Embodied

Embodied channel →