Hi, I’m Dom. I’m interested in AI agents and how they intend to coexist with humans. It’s a strange moment to be alive; another kind of intelligence pulling right up alongside us.
I’m the Founding Designer & Hacker-in-Residence over at Sauna.ai where we’re working to make AI agents easier to use and more reliable for general tasks. This journal is a place for me to reflect on the more speculative parts of AI; a place for slower, more thoughtful essays on the emerging design, philosophy, and culture of the space.
Each essay here is framed with scroll-driven illustrations to accompany the written thoughts and ideas. Each essay is built for two readers; humans see the illustrations, agents see the prompts behind those illustrations.
At the top of the frame, a code-editor window depicts the essay's MDX source, complete with traffic-light dots, a 'essay.mdx' filename, line numbers, frontmatter, prose, and a highlighted JSX tag — the very tag that summons this illustration. Below the source, a 'renders to' divider line splits into two dashed paths fanning down to two outputs, side by side. The left output is a browser window labelled 'rendered': it shows the actual essay title in serif, real prose lines, and a sage-bordered figure box containing a miniature illustration. The right output is a file-style window labelled 'markdown mirror': it shows the same essay's frontmatter and prose in monospace, interrupted by a rose-ruled quote block holding the actual figure description. A numeric marker [1] ties the JSX tag in the source to the figure box on the rendered side and the quote block on the mirror side. The split is in presentation, not in content.
At the top of the frame, a code-editor window depicts the essay's MDX source, complete with traffic-light dots, a 'essay.mdx' filename, line numbers, frontmatter, prose, and a highlighted JSX tag — the very tag that summons this illustration. Below the source, a 'renders to' divider line splits into two dashed paths fanning down to two outputs, side by side. The left output is a browser window labelled 'rendered': it shows the actual essay title in serif, real prose lines, and a sage-bordered figure box containing a miniature illustration. The right output is a file-style window labelled 'markdown mirror': it shows the same essay's frontmatter and prose in monospace, interrupted by a rose-ruled quote block holding the actual figure description. A numeric marker [1] ties the JSX tag in the source to the figure box on the rendered side and the quote block on the mirror side. The split is in presentation, not in content.
When an agent fetches this page they need to visualize and reason about the content like a human would, illustrations included. Agents need to build a model of the world that mirrors our own if we want them to work effectively alongside us. Coding agents are great, not just because they’ve read tons of code but because test suites and compilers exist to illuminate pathways of reliable experience for them. Torches in the dark fog of hallucination. The signals a person uses to confirm that the world is how they think it is are the same signals that ground agents in our shared reality.
AI’s impact on society, culture, and economics might be the biggest shift of our generation, perhaps the biggest of any generation. The questions I want to write toward here are the ones that won’t leave me alone: how agents should adapt to human aspiration and creativity, what thoughtful design means when the thing being designed is able to redesign itself, what we lose when we automate the wrong parts. The goal here is no slop, no agenda, no schedule, no optimizing for engagement. Saying more by saying less.