human interaction

  • Running Local AI on a 2017 iMac: What Actually Happened

    The iMac That Couldn’t If you’ve been searching for honest, real-world results on running a local AI setup on older hardware, specifically installing Ubuntu on a 2017 iMac and running an open-source AI agent locally, this post is for you. No sponsored results. No, “It worked great!” YouTube thumbnails. Just what actually happened when I…

  • From SSH Friction to a Private LLM Cloud: Building a Private, Repeatable Ollama + EC2 Setup

    Setting up a private, cloud-powered LLM environment on AWS sounds straightforward, until you actually do it. What started as a simple goal… “run Agent Zero with Ollama on EC2 without exposing sensitive data” …quickly turned into a maze of SSH quirks, Docker inconsistencies, networking gotchas, and AWS nuances. The good news: once you get through…

  • Designing AI Systems That Actually Work

    I’ve been designing and deploying custom AI agents, not generic tools, but systems built around how individuals and businesses actually work.

  • The Hard Problem No One Is Solving in AI: Identity Continuity

    There’s a quiet assumption baked into most conversations about AI right now: If you can export the data, you can recreate the experience. That assumption is wrong. And the gap between those two things, data and experience, is where the real problem lives. This Isn’t About Chat History Most people approaching “AI migration” are thinking…