Solo architect of self-modifying AI systems.
Building agents that read their own code, fix their own bugs, and evolve their own architecture.
I design and build Genesis Agent — a self-aware, self-modifying cognitive AI agent with a 12-phase boot system, hexagonal architecture, and organism-level autonomy. It doesn't just use LLMs — it is one, wrapped in 237 modules of self-verification, self-repair, and autonomous planning.
237 source files · 80k LOC · 3311 tests · 142 services
12 boot phases · 0 cross-layer violations · 100% architectural fitness
348 typed events · 85 @ts-ignore (down from 336) · TSC clean (0 errors)
Runs on Claude, GPT-4, Kimi, DeepSeek, or local models via Ollama
┌─────────────────────────────────────────────────────────────┐
│ GENESIS AGENT — Cognitive Architecture (v7) │
├─────────────────────────────────────────────────────────────┤
│ │
│ ┌──────────┐ ┌──────────┐ ┌──────────┐ ┌──────────┐ │
│ │Foundation │→ │Intellig. │→ │Capabilit.│→ │Planning │ │
│ │ Phase 1 │ │ Phase 2 │ │ Phase 3 │ │ Phase 4 │ │
│ └──────────┘ └──────────┘ └──────────┘ └──────────┘ │
│ ┌──────────┐ ┌──────────┐ ┌──────────┐ ┌──────────┐ │
│ │Hexagonal │→ │Autonomy │→ │Organism │→ │Revolution│ │
│ │ Phase 5 │ │ Phase 6 │ │ Phase 7 │ │ Phase 8 │ │
│ └──────────┘ └──────────┘ └──────────┘ └──────────┘ │
│ ┌──────────┐ ┌──────────┐ ┌──────────┐ ┌──────────┐ │
│ │Cognitive │→ │ Agency │→ │Extended │→ │ Hybrid │ │
│ │ Phase 9 │ │Phase 10 │ │Phase 11 │ │Phase 12 │ │
│ └──────────┘ └──────────┘ └──────────┘ └──────────┘ │
│ │
│ AwarenessPort — lightweight interface (replaces Phase 13) │
└─────────────────────────────────────────────────────────────┘
| System | Purpose |
|---|---|
| CognitiveWorkspace | 9-slot working memory (Global Workspace Theory) |
| OnlineLearner | Reactive learning from every interaction |
| LessonsStore | Cross-project persistent learning (~/.genesis-lessons/) |
| DreamCycle | Offline memory consolidation + insight generation |
| AwarenessPort | Lightweight self-awareness interface (replaced 14-module Consciousness Layer in v7) |
| PreservationInvariants | 11 hash-locked semantic safety rules |
| ColonyOrchestrator | Multi-agent goal decomposition + merge |
| DeploymentManager | Strategy-based live deployment with rollback |
| DisclosurePolicy | Trust-based information sovereignty + social engineering defense |
| DaemonController | External control via Unix Socket / Named Pipe |
Solo developer. I ship iteratively — build, verify, release. Every module has tests. Every event has a payload schema. Every release runs through 5 CI gates: tests, architectural fitness, event audit, event validation, and type checking. The codebase maintains itself.
Stack: JavaScript · Node.js · Electron · MCP Protocol · Ollama · Anthropic · OpenAI · Kimi · DeepSeek
Germany · Building autonomous systems that understand their own architecture.
