The Forge

The reliable, always-on foundation for your sovereign cluster. Pre-configured with Proxmox and Aviary OS to act as your primary control plane.

A localized, air-gapped AI inference engine wrapped in rugged leather and brass. Absolute privacy you can carry with you.

The physical memory core of the Fin-Hawk system. A bare-metal server engineered specifically for massive vector databases and high-speed RAG.

A compact, low-power local AI inference engine designed specifically to fit and run inside your Tesla Model Y.

An ultra-low-power mobile inference engine powered by Texas Instruments' AM68A Edge AI vision processor and integrated NPU.

The absolute boundary between your sovereign data and the outside world. A bare-metal pfSense gateway.

The highly-efficient, low-power Kubernetes (K3s) control plane for the Fin-Hawk system.

A compact, ultra-efficient edge node powered by Ryzen AI for fast-path reasoning and voice presence.

The heavy-duty data backbone of your cluster. A high-resource Dell server for orchestration, vector databases, and massive context ingestion.

The sole active physical node available. A specialized inference engine built on a desktop chassis leveraging late generation GPUs to provide robust vector database serving and context servers.

The unsung hero of the cluster. A low-power, battery-backed node dedicated to monitoring and backups.

A complete infrastructure-as-code blueprint for deploying a multi-GPU sovereign AI cluster running Aviary OS.

The brain of the Fin-Hawk system. Deploy a dedicated Model Context Protocol (MCP) server for instant knowledge retrieval across your autonomous agents.

The exact Ansible playbooks and pfSense configurations used to securely route and air-gap the Fin-Hawk AI cluster.

A comprehensive deployment guide for linking multiple local inference nodes into a unified AI swarm using Ray and FastMCP.

Build your own portable, air-gapped AI inference engine. Complete parts list, CAD files, and software loadout.

The complete DIY guide for building and integrating a private, local AI inference engine into your Tesla Model Y.