Dosidicus

🦑 A transparent cognitive sandbox disguised as a digital pet squid with a neural network you can see thinking

Micro neural engine for small autonomous agents that learn via Hebbian dynamics and grow new structure

• Part educational neuro tool, part sim game, part fever dream
• Build-your-own neural network - learn neuroscience by raising a squid that might develop irrational fears
• Custom simulation engine using Numpy - No Tensorflow or PyTorch
• Most AI is a black box; Dosidicus is transparent - every neuron is visible, stimulatable, understandable.
• Starts with 8 neurons — grows via neurogenesis and rewires using Hebbian learning.

Dosidicus is a digital squid born with a randomly wired brain.

Feed him., stimulate neurons, watch him learn.

• He starts with 8 neurons.
• He rewires through Hebbian learning.
• He grows new structure via neurogenesis.
• He forms memories.
• He develops quirks.

Every squid is different. Every save file is a cognitive history.

Custom simulation engine:

• Built from scratch in NumPy
• No TensorFlow. No PyTorch.
• Fully visible neuron activations
• Structural growth over time
• Dual memory system
• Headless training mode
• Most AI is a black box: Dosidicus lets you see the mind forming.

Want the full conceptual philosophy behind Dosidicus? → Read the Cognitive Sandbox Manifesto