GitHub Repository

A Bio-Mimetic Digital Organism . Unlike static AI, Genesis feels pain, gets bored, sleeps, and evolves its own code using Liquid State Machines. Exploring the future of Synthetic Consciousness.

0 starsPython

Project Genesis – A Bio-Mimetic Digital Organism Using LSM

Name: Project Genesis – A Bio-Mimetic Digital Organism Using LSM
Availability: InStock
Author: Jeevan_Joshi

by Jeevan_Joshi·Feb 11, 2026·1 point·0 comments

Visit Project View on HN

AI Analysis

●●SolidWizardryRabbit HoleBig Brain

2,100-neuron LSM with dopamine loops is clever biology sim; unclear if it does anything practical.

Strengths

•Liquid State Machine + Hebbian learning + hormone simulation (cortisol, dopamine, oxytocin) is genuinely novel architecture—not an LLM wrapper
•Numba-accelerated neuroplasticity + sleep-cycle memory consolidation shows deep technical craft
•100% local, no API dependencies, with biologically-grounded constraints (pain increases noise, boredom triggers questions)

Weaknesses

•No clear use case or benchmark; 'feels pain' and 'gets bored' are evocative but unvalidated claims
•16 stars suggests niche interest; no evidence of reproducibility or downstream application
•README conflates marketing hype ('exploring synthetic consciousness') with actual capabilities; unclear what it actually *does*

Post Description

I built a local digital organism using a Liquid State Machine architecture (2,100+ neurons). It simulates biological processes like dopamine decay, cortisol-based stress (injecting noise into the reservoir), and Hebbian learning/neuroplasticity using Numba. No LLM APIs used. Code is 100% Python.