AI Platform Engineer | Founder, West AI Labs
Building production AI infrastructure that enterprises can deploy, govern, and trust.
I build AI systems that run like enterprise software, not experiments. My focus is local-first AI platforms — models on hardware you own, every action logged and governed, data that never leaves your building.
Before founding West AI Labs, I spent 25+ years shipping production systems at O'Reilly Auto Parts, Bass Pro Shops, IBM, Bank of America, and AIG — from multi-datacenter infrastructure to enterprise middleware to global financial platforms.
Today I use MCP (Model Context Protocol) and Claude Code as daily production tools. My AI agent spawns sub-agents, routes to different models by complexity, and writes its findings to files I review. That's not a chatbot — that's infrastructure.
US Navy veteran 🇺🇸
A modular, local-first AI platform for inference, orchestration, and agent hosting:
| Project | Description |
|---|---|
| nebulus-prime | GPU inference runtime for Linux/NVIDIA (TabbyAPI, ExLlamaV2, vLLM) |
| nebulus-edge | Apple Silicon AI platform (MLX inference, agent hosting) |
| nebulus-gantry | Self-hosted AI chat interface with RAG, memory, and admin controls |
| nebulus-core | Shared library across Linux and macOS platforms |
| nebulus-atom | Autonomous AI agent CLI with task orchestration and tool governance |
| nebulus-forge | AI-native project scaffolder with built-in governance and quality gates |
AI/ML: MCP, Claude Code, TabbyAPI, ExLlamaV2, MLX, vLLM, Ollama, ChromaDB
Backend: Python, FastAPI, Java, Node.js
Frontend: React, TypeScript, Vite
Infrastructure: Docker, Ansible, systemd, NVIDIA Container Toolkit
Practices: Conventional Commits, pre-commit hooks, pytest, Gitflow
- 🌐 westailabs.com
- 🐦 X/Twitter
The demo is easy. Production is hard. I do the hard part.