A signal becomes
a number.

Everything that follows starts here: something messy and physical gets written down as an array of floats.

Then the numbers get weights.

Weights are just how much each number counts. Training is the search for better ones.

And the weights start predicting.

Seven working models are below. Break them and watch what happens.

Scroll
Running entirely in this tab

Machine learning you can put your hands on

No backend, no API keys, no waiting. Gradient descent, K-Means, A*, backpropagation, attention, sorting and packet switching all execute as plain JavaScript a few centimetres from your cursor.

Lab 01

You versus gradient descent

Plot some points, then try to drag the line onto them by hand. When you give up, let the optimiser do it and watch how it gets there.

Telemetry

Data points
0
Slope w
0.00
Intercept b
150
Mean squared error
—
Click to add a point · residuals shown in pink

Clustering

K-Means finds groups nobody labelled. Add points and watch the centres chase them.

Clusters
Click to inject a point

Pathfinding

A* expands the cheapest guess first. Pink is explored, cyan is queued.

path — expanded —
Drag to draw walls

Forward pass

Signal moves left to right; thickness is weight magnitude, colour is sign.

epoch 0
loss 1.000

Attention

Each word scores every word before it. Write your own sentence and see.

 

Sorting

Compare, swap, repeat. Bubble sort floats the biggest bar to the right each pass. Try the others on the same array.

Algorithm
compares 0 swaps 0 average case O(n²)
Drag across the bars to draw your own array

 

Star topology

Every device has its own cable to one central box. Pick a sender, then a receiver, and follow the frame through the middle.

Devices Centre
frames 0 wasted copies 0 MAC table 0/6
Click a device to send · click another to receive

 

Nine ways to think

Each topic gets its own page with a working demo, the maths that drives it, and where it actually shows up. Poke the chip if you want the grid to notice you.

Search & Symbolic AI 1 Interactive Concepts

Artificial Intelligence

From 1956 Dartmouth Dartmouth logic to heuristic state spaces, minimax game trees, and knowledge representation.

Foundations & Architecture 2 Interactive Concepts

Core Computer Science

TCP/IP packet routing, dynamic SPF graph traversal, asymptotic Big-O analysis, sorting algorithms, and heap data structures.