Nicheloom

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

Neural Particle Automata

Details

External ID
48642030
Source
HN
Company
—
Product
Neural Particle Automata
Website domain
github.io
Launched
June 23, 2026
Cohort
—
Upvotes
88
Upvotes percentile
0.9016393442622951
Tags
—
Fetched at
Sept. 7, 2026, 9:26 p.m.
Updated at
Sept. 7, 2026, 9:26 p.m.

Description

Neural CAs model self-organizing pattern formation on grids. Now the grid is gone. Each cell is an agentic particle that can move freely in space and change its state.While each particle follows a simple shared rule, many together can grow complex morphologies or form intricate patterns. The resulting particle system as a whole can regenerate from damage and exhibits surprising emergent behavior.Try cutting the lizard and watch it heal itself!

Enrichment

Theme
3D graphics and physics simulation tools
Vertical
Horizontal
Function
Model & infra
Audience
Developer
AI stance
AI-native
Project type
Hobby / open-source project
Normalized one-liner
neural particle automata framework
Manually corrected
False

Could you build this?

No This is academic research accepted at SIGGRAPH featuring differentiable Smoothed Particle Hydrodynamics (SPH) and custom memory-efficient CUDA kernels.

What it would actually take: The system requires implementing Lagrangian particle dynamics using differentiable SPH operators formulated inside PyTorch or JAX, accompanied by custom CUDA kernels for neighbor search and dynamic spatial hashing. The primary challenges are achieving gradient stability through chaotic particle trajectories, managing $O(N)$ spatial neighborhood queries under tight GPU memory limits, and designing learnable cellular update rules that generalize. It demands elite expertise in computational physics, GPU shader/kernel programming, and scientific machine learning.

Discussion

19 comments analyzed.

Competitors mentioned: cells2pixels, Michael Levin's electrical communication research, self-organizing textures (Distill)

Concerns raised: Particles only grow reliably from egg-like initial conditions, Switching rules mid-rollout produces messed up morphology, Can be disrupted enough that patterns can't re-form

Feature requests: 3D point cloud/gaussian splat experiments with video demo, Cell division/splitting capability, Texture synthesis using similar approach, Pattern persistence when selecting new patterns without refresh

Competitors

Other products that read as similar to this one — 36 launches clear the similarity bar, closest 8 shown.

Attention rank: #3 of 37 (itself plus its competitors, highest first — normalized so YC and Product Hunt are compared fairly).

Launched 177 days after the earliest competitor.

Other launches for this product

Same idea, different domain

Nobody's really built a model & infra tool for Fintech yet.