Skip to content
View moonrunnerkc's full-sized avatar

Block or report moonrunnerkc

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
moonrunnerkc/README.md
Brad Kinnard Banner

Brad Kinnard

AI Architect | Founder @ Aftermath Technologies | Offline Systems Engineer


I build autonomous cognitive architectures that run fully offline. My systems don't just process data; they possess beliefs that compete, mutate, and evolve based on evidence. Real understanding grown from hard constraints, not endless data.


Website Email LinkedIn X


⚡ Engineering Stack

I prioritize tools that allow for local execution, strict verification, and high performance.

Core Logic Infrastructure Cognitive Design
Python TypeScript Docker Linux Belief Ecologies
PyTorch Offline LLMs Redis FastAPI RL Pipelines
Multi-Agent Systems Async Architectures Immutable Memory

🛠 Active Development

Current focus: Offline cognitive operating systems with persistent, evolving intelligence.

The Kernel. An offline cognitive OS with signed logs, RL goal evolution, and automatic failure recovery.

  • Focus: Audit trails, contradiction resolution, anomaly detection.

The Mind. A multi-agent belief system where ideas live or die based on utility.

  • Focus: Dynamic memory, conflict detection, mutation, decay.

The Evolution. Pure tension-driven development to prevent reward hacking.

  • Focus: Evolution without fine-tuning drift.

The Workforce. Parallel local coding sessions with isolated branches.

  • Focus: Smart dependency handling, swarm coordination.

The Search. Local retrieval fusion tuned by reinforcement learning.

  • Focus: Combining RAG, graphs, and caches without the cloud.

I am always open to serious collaboration on offline cognitive architectures and verifiable AI.

© 2026 Bradley R. Kinnard / Aftermath Technologies

Pinned Loading

  1. adaptive-belief-ecology-system adaptive-belief-ecology-system Public

    ABES: Adaptive Belief Ecology System. It is a dynamic AI memory, not static, where beliefs live, decay, conflict, and mutate like organisms. 15 agents and RL run this self-updating cognitive ecosys…

    Python 1

  2. shadow-ecology shadow-ecology Public

    A self-evolving cognitive architecture that grows a real mind through tension, contradiction, and time. No reward model. No fine-tuning. No alignment tax. Only decay, mutation, and the slow pressur…

    Python

  3. code-style-enforcer code-style-enforcer Public

    A cloud-ready, multi-agent code analysis system that runs style, naming, documentation, and security checks in parallel and adapts to your preferences using reinforcement learning. Designed for fas…

    Python 1

  4. copilot-swarm-orchestrator copilot-swarm-orchestrator Public

    Parallel execution of GitHub Copilot CLI sessions with dependency-aware scheduling, verification, and per-agent git branches.

    TypeScript 11

  5. rlfusion-orchestrator rlfusion-orchestrator Public

    Local RL-driven AI orchestrator fusing RAG, CAG, and graphs - Built offline on consumer hardware.

    Python

  6. aasms aasms Public

    🔄 Local-first AI system that evolves its own codebase through adversarial Red/Blue team dynamics. Ollama-powered, Docker-isolated, with cryptographic safety guards. Research prototype exploring rec…

    Python