Nicheloom

Market intelligence for builders — see what's gaining traction before it's crowded.

Decided to play god this morning, so I built an agent civilisation

Details

External ID
47195530
Source
HN
Company
—
Product
Decided to play god this morning, so I built an agent civilisation
Website domain
github.com
Launched
Feb. 28, 2026
Cohort
—
Upvotes
51
Upvotes percentile
0.8194070080862533
Tags
—
Fetched at
Sept. 7, 2026, 9:25 p.m.
Updated at
Sept. 7, 2026, 9:25 p.m.

Description

at a pub in london, 2 weeks ago - I asked myself, if you spawned agents into a world with blank neural networks and zero knowledge of human existence — no language, no economy, no social templates — what would they evolve on their own?would they develop language? would they reproduce? would they evolve as energy dependent systems? what would they even talk about?so i decided to make myself a god, and built WERLD - an open-ended artificial life sim, where the agent's evolve their own neural architecture.Werld drops 30 agents onto a graph with NEAT neural networks that evolve their own topology, 64 sensory channels, continuous motor effectors, and 29 heritable genome traits. communication bandwidth, memory decay, aggression vs cooperation — all evolvable. No hardcoded behaviours, no reward functions. - they could evolve in any direction.Pure Python, stdlib only — brains evolve through survival and reproduction, not backprop. There's a Next.js dashboard ("Werld Observatory") that gives you a live-view: population dynamics, brain complexity, species trajectories, a narrative story generator, live world map.thought this would be more fun as an open-source project!can't wait to see where this could evolve - i'll be in the comments and on the repo.https://github.com/nocodemf/werld

Enrichment

Theme
indie hacker passion projects
Vertical
Horizontal
Function
Agent / copilot
Audience
Developer
AI stance
AI-native
Project type
Hobby / open-source project
Normalized one-liner
multi-agent simulation environment
Manually corrected
False

Could you build this?

Partial While an agent sim visualizer is easy to code, simulating emergent language and social evolution from blank neural networks requires complex reinforcement learning and evolutionary algorithms.

What it would actually take: The system requires a fast simulation environment (often Rust, C++, or Python with JAX/PyTorch) executing neuroevolution (NEAT) or multi-agent reinforcement learning (MARL) where agents interact via perceptual inputs and action spaces. The hard part is designing reward functions, environmental pressures, and evolutionary dynamics that produce genuine emergence rather than extinction or random noise. It requires deep research expertise in computational biology, multi-agent RL, and evolutionary computation.

Discussion

20 comments analyzed.

Competitors mentioned: Polyworld, SharpNEAT, Conway's Game of Life

Concerns raised: Simulation feels cyclical after running for a while, Hard to detect patterns and emergent behaviors in agents, Difficult to identify preconceptions or biases in evolved agents

Feature requests: Time-lapse visualization of world evolution, Pattern detection analysis over agent positions/actions, Animation showing world history/changes over time

Competitors

Other products that read as similar to this one — 205 launches clear the similarity bar, closest 8 shown.

Attention rank: #42 of 206 (itself plus its competitors, highest first — normalized so YC and Product Hunt are compared fairly).

Launched 120 days after the earliest competitor.

Other launches for this product

Same idea, different domain

Nobody's really built a agent / copilot tool for Agriculture yet.