Corvus
A persistent personal AI agent — semantic memory on local Ollama embeddings, hybrid Python + LLM execution, and cron-driven daily briefings. Privacy-first, runs on a Mac Mini.
I'm Mike Jones — 29 years building systems that ship, now helping organizations turn fast-moving AI into infrastructure they can actually run. Not a demo. Daily dependable.
Microsoft · Xbox / Kinoo / 8 Circuit Studios / Velocity Partners

Across 29 years I've led teams of 120+, managed $12M+ budgets, shipped 6 AAA launches, and hold a patent from the original Xbox SDK — experience I now point at making AI real inside organizations.
I help people and organizations operate at the intersection of AI, autonomy, and resilience.
Consulting
Making AI actually work inside organizations.
Fractional PMO + AI implementation for teams that need AI to move from demo to dependable. Built on AAPD — AI-Augmented Process Design.
Teaching
AI & Claude Code courses for senior career-changers.
A teaching partnership with Ron Nash's GHN Academy — helping directors and VPs build AI fluency and reposition their expertise. Course series launching July 2026.
Publishing
Practical resilience for people exiting the default.
A publication and growing community on building parallel systems — 2,100+ subscribers, organized around a 7 Pillars framework. Top essay: "7 Steps to Quietly Exit a System" (1,000+ likes).
Product
Your career, as an interactive AI.
An AI SaaS that turns your career into an interactive chatbot — a way for a recruiter or hiring manager to get to know you before the first call. It grew out of my own frustration with a portfolio that buried the answers people actually wanted. The assistant on this page is powered by it.
AI as executive function
I'm building — and living in — a system that externalizes memory, makes time visible, and reduces the friction between intention and action. Corvus captures and triages; daily briefings surface the one next thing; AI handles the mechanics so attention goes where it matters. It's resilience at the scale of a single mind, and it's becoming a framework I want to teach.
How I think about itA persistent personal AI agent — semantic memory on local Ollama embeddings, hybrid Python + LLM execution, and cron-driven daily briefings. Privacy-first, runs on a Mac Mini.
A cross-AI JSONL memory layer I built because no tool had one — Claude had no persistent memory, ChatGPT's couldn't be exported, my local model's couldn't be reached. A shared ledger (plus a validation agent) gives all three persistent, shared context.
A privacy-first, self-hosted LLM stack — Ollama, OpenWebUI, and an MCP bridge, auto-starting and always on.
Consulting, a course for your team, or a straight conversation about where AI actually fits.