About this site

Every major technology I’ve lived through has given me something. Computers gave me a creative outlet. The internet gave me access to everything. Smartphones... honestly, I’m still not sure what smartphones gave me besides screen time. But LLMs are on another level. The amount I’ve been able to build, learn, and ship since I started working with them is unlike anything I’ve experienced before. It genuinely feels like I’ve gained hours back in my day, both professionally and personally. Wozniak once joked he’d invent a pill where one hour of sleep works like eight. This is the closest anything has come.

That’s not just because the models are good. It’s because of what you can build around them. I’ve spent the last few years working on AI products, agents, pipelines, prototypes, and the thing that keeps pulling me in is the layer between the model and the user: prompting strategies, scaffolding patterns, context engineering. The stuff that turns a base model into something you can actually ship. The model is the engine, but the harness is what makes it go somewhere.

This site is where I work through those ideas in public. Deep dives into the research, experiments testing whether it holds up in practice, interactive visualizations that try to make the concepts click. I explain things the way I wish someone had explained them to me: honestly, visually, and without assuming you already know the field.

I’m writing this primarily for myself. It’s how I learn, by forcing myself to explain things clearly enough that someone else could follow along. It also saves me from repeating myself: I spend a lot of time helping colleagues, friends, and family get more out of these tools, and at some point it’s easier to just write it down. If you’re building with LLMs and trying to figure out how to get more out of them, I hope you find something useful here too.

— Max