Tools · Experiments

Learn by measuring something real

Experiments are guided, hands-on labs that run in your browser with an OrbitCore board. You do not start from a ready-made dataset. You build the instrument, collect the data yourself, and follow it all the way to a working machine-learning model.

Each one is written for a mixed room: engineers, biologists, students and people who are simply curious. Every step explains what is happening and why.

OrbitCore development board

The framework

Every experiment follows the same five stages

1

Build

Describe what to measure. Physical Vibe Coding writes the program and you flash it to OrbitCore.

2

Collect

Record real signals live in the browser and label them as you go.

3

Clean

Filter the noise and turn raw readings into meaningful features.

4

Learn

Train and compare machine-learning models on the data you made.

5

Predict

Test the model on new data, then run it live on your own signals.

Available now

Choose an experiment

Experiment 01

Build Your Own Biological Dataset

Workshop: Introduction to Machine Learning in Bio-Mechatronics & Bioinformatics

Put your fingertip on a MAX30100 pulse sensor. Record yourself at rest, during light activity and in recovery, train a model to tell the three apart, then put the model on OrbitCore so the board predicts by itself and blinks the answer.

🫀 MAX30100 pulse sensor💡 Infrared + red lightRandom Forest🔴 Runs on the boardAbout 45 minBeginner friendly
Open experiment →

Experiment 02 · A

Resistance Lab: the Salt Water Detector

Workshop: Introduction to Machine Learning in Bio-Mechatronics & Bioinformatics

Build two Wheatstone bridges on OrbitCore: one probe stays in plain water as a control, the other tests the sample on A2/A3. Set a threshold on the difference between the bridges, test it on real cups and store it on the board, which answers with its LED. No machine learning.

🌉 Two Wheatstone bridges💧 Water control on A0/A1📏 Threshold rule🔴 Runs on the boardAbout 25 minNo ML
Open experiment →

Experiment 02 · B

Resistance Lab: Teach a Model

Workshop: Introduction to Machine Learning in Bio-Mechatronics & Bioinformatics

The same two bridges, now with machine learning. Record labelled samples with notes (tap water, salt water, or your own), train a model on how the sample bridge differs from the water bridge, predict live, then put the model on OrbitCore.

🌉 Two Wheatstone bridges🏷️ Your own labels + notesRandom Forest🔴 Runs on the boardAbout 40 min
Open experiment →

More experiments will be added here as new workshops run.