Hi, I'm Fred.

I've spent the last few years building AI systems inside companies. I started TinyMacro to work on hard problems where physical and economic limits decide the answer.

What I think matters

I think AI is the most important technology of my lifetime so far, and I enjoy putting it to work inside companies. But I have loved nature my entire life, and I don't feel a real pull toward any work unless it serves something I care about that much.

The resources we depend on are finite. Water, soil, fisheries and forests come back only as fast as natural processes allow. As AI and the rest of our technology grow, so does the demand on them. I want to build around that limit.

Most talk about AI is about automating knowledge work. I am more interested in using it to look after the natural world, and to understand what these systems can provide and how to help them provide more. We already collect a huge amount of data about the Earth, and most of it sits unused.

Companies pollute far more than individual consumers do, so that is where I want to work. Large companies decide how much energy they use, where it comes from and where they build.

Energy is the clearest case. Every new data center needs grid capacity, and a grid connection is itself a limited resource. When a project sits in a queue or withdraws from it, the capacity and the money behind it go unused.

My current research is MacroLayer, starting with power for large loads: data centers and industrial sites deciding where to build, whether to wait for a grid connection or generate on site, and how to buy clean power. Those outcomes are public, so a plan can be checked against what actually happened.

Experience and education

For the last few years I have done advisory AI work for companies, and in that time I built AI infrastructure for Wheel the World, an accessible travel company, and for Larta Institute, a federal research accelerator working with NOAA and USDA programs. I still take that work through TinyMacro LLC, where I built an agent harness for ShoplyAI's SEO, content and outbound, and where I am now doing AI implementation for a law firm.

At TinyMacro I build harnesses: agents on Cloudflare Durable Objects that work through a task and check the result. MacroLayer is designed so that a classifier proposes a claim, code checks it against its sources and a person reviews it before it counts as a finding.

I am finishing an economics degree at UCLA in December 2026, with a minor in Environmental Systems and Society.

Get in touch