A few things I have come to believe from building with AI over the past few years, from inside an early Microsoft partnership to running models on my own hardware.
In 2022 I was leading our partnership with Microsoft as a Copilot development partner. I spent that year in regular conversations with Microsoft leaders, exploring what extensibility of Copilot would look like and how you would orchestrate many AI tools together. It kept resolving to one idea. Instead of a single model doing everything, you build agents, each responsible for a nuanced part of a value chain. A real workflow is not one call to a model. It is dozens, and more likely hundreds, of agents working in parallel, each owning a narrow job. That is where this is all headed.
Not one model doing everything, but hundreds of agents, each owning a narrow job.
The models keep improving, and quickly. We have reached the point where knowing how to build with AI is just as important as understanding the nuanced workflows you are asking agents to run. You need both halves: the domain knowledge to define the job precisely, and the craft to build the agent that actually does it well. Teaching that craft, on real work, is what Maai Services is for.
At the same time we were meeting models like ChatGPT for the first time, it became obvious that data centers, infrastructure, and energy would shape how the entire AI economy gets built. So in 2023 I started experimenting with Apple silicon and moved my work onto it. Today I run my businesses and my agents on Mac minis, with a cluster of Mac Studios coming, using a mix of cloud and local models. Owning the hardware means private, always-on AI that does not depend on anyone else’s uptime or terms. I help others do the same through Maai Machines.
Owning the hardware means private, always-on AI on your terms.
Put together, these point one way: toward people and small companies who can build with AI themselves, orchestrate their own agents, and run them on infrastructure they control.