Skip to content
View Dominien's full-sized avatar

Block or report Dominien

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Dominien/README.md

Marco Patzelt

I build worlds where AI agents figure things out on their own.

Contact Location


What I'm building

Brunnfeld — A medieval village economy that runs itself. 20 LLM agents with no behavioral instructions, no trading strategies, no goals. Just hunger, tool degradation, seasonal yields, and a marketplace. They negotiate prices, extend credit, hoard goods, and starve if they don't figure it out. Playable viewer with pixel art map, God Mode events, and a join as villager mode. Stars

TypeScript · Any LLM · Free models supported · The engine is the personality

Built on top of:

Social Agent Sim — The project that started it all. 6 LLM agents in a Berlin apartment building with zero personality prompts. Just a 2 line bio, physics constraints, and 15 possible actions. Identity, relationships, and collective action emerge from the architecture. Brunnfeld took this social layer and added a full economy on top. Stars


Open source

Project What it does Stars
Brunnfeld 20 agent medieval village economy. No instructions, just world physics. Playable with God Mode. Stars
Social Agent Sim 6 agent social simulation in a Berlin apartment. The foundation Brunnfeld was built on. Stars
Agentic Orchestration Layer Cognitive middleware with runtime code generation and self correcting triangulation. Stars

What I do professionally

Integration Engineer — Real time middleware between enterprise CRMs (Microsoft Dynamics 365) and modern frontends (Webflow, Next.js). Entity reconciliation, automated content pipelines, Redis caching. Production systems handling live business data daily.


Stack

AI — Claude, Gemini, OpenRouter, Supabase (PostgreSQL + pgvector), E2B Sandboxing
Backend — Node.js, TypeScript, Vercel Serverless, Redis / Upstash, Docker, Dynamics 365 API
Frontend — Next.js, Tailwind, Zod, SSE streaming, Canvas rendering
Payments & Auth — Stripe, Google OAuth, Supabase RLS


Environment design over prompt engineering.

Pinned Loading

  1. brunnfeld-agentic-world brunnfeld-agentic-world Public

    Medieval village economy simulation — up to 1000 LLM agents across multiple villages, driven by supply chains, hunger, tool degradation and market pressure.

    TypeScript 112 15

  2. claude-code-commander claude-code-commander Public

    Desktop command center for managing multiple Claude Code sessions across codebases. Multi-agent orchestration via MCP.

    TypeScript 14 2

  3. social-agent-sim social-agent-sim Public

    Multi-agent social simulation — 6 AI agents in a Berlin apartment building discover identity through lived experience

    TypeScript 9 2

  4. agentic-orchestration-layer-model agentic-orchestration-layer-model Public

    A System 2 Cognitive Middleware that replaces static data pipelines with runtime code generation. Features Gemini 3 Pro, E2B Sandboxing, and Self-Correcting Triangulation.

    TypeScript 6