Who is Amelia Wattenberger and what does she work on?

Amelia Wattenberger is a Principal Research Engineer at GitHub, where she works on a small R&D team exploring novel user interfaces and machine learning. Her personal site, wattenberger.com, is a working notebook of that thinking: a mix of essays on interface design, data visualization tutorials, and experiments in AI tools, generative art, and laser-cut fabrication. If you're looking for practical D3 or React guidance, her earlier writing is the most directly useful part of the site; if you're interested in where interfaces and AI are heading, the newer notes and project list are the place to start.

What she does at GitHub

Her current role is Principal Research Engineer on a small R&D team within GitHub. The site describes the work as "exploring novel UIs + playing with ML." One named project is Code Atlas, framed around making LLM responses more robust and easier to understand. That framing is consistent with the rest of her public work: the recurring question is how to make complex systems — data, code, model output — legible to the people using them.

Recurring themes

Her writing and projects cluster around a few durable interests:

• Data visualization — D3, interactive charts, force simulations, visualizing codebase changes over time, visualizing time as a spiral.
• Front-end craft — React, SVG, the CSS cascade, hooks, scaling SVG elements.
• AI as a tool for thought — writing and brainstorming tools, embeddings, moodboarding with Stable Diffusion, generating a font with text-to-image models.
• Interface design — infinite canvases, "chat is AI's command line," "our interfaces have lost their senses," fish-eye text.

The through-line is interfaces that help people think, rather than interfaces that just display.

Representative writing

Her "Previous writings" list is the most concrete, reusable part of the site. Titles include:

If you're learning D3 or trying to combine it with React, these are the pieces to read first — they address the specific friction points (who owns the DOM, how to structure updates) that come up in that combination.

Current experiments and projects

The "currently noodling on" and "what I've been building" sections read as an active lab notebook rather than a portfolio. Recurring entries include:

• AI writing and brainstorming tools — one described as a brainstorming tool for organizing thoughts, another as an AI writing tool that helps you improve, and a precursor that helps "sculpt ideas like clay."
• Infinite canvases — an essay plus ongoing design thinking about how to improve them.
• Data viz + AI — combining text-to-image with data visualization, moodboarding with Stable Diffusion.
• Generative and physical art — generating images from laser-cut kumiko patterns, wall art larger than her laser bed, hacking undo/redo into a keyboard.
• JavaScript tooling — visualizing which JS tools people use together, exploring common structures across JS frameworks.

Some entries link to "Read more"; others are just a line of status. Treat the list as a signal of what she's actively thinking about, not as finished products.

The book and course

She is the author of Fullstack D3 and Data Visualization, available as a book and course. It's the most structured, end-to-end resource she's produced, and the natural next step if the shorter D3 articles leave you wanting a full curriculum.

How to use the site

• Learning D3 or React+D3: start with the "Previous writings" list, then the book/course.
• Interface and AI design thinking: read the "I'm thinking about..." notes — they're short, opinionated prompts rather than tutorials.
• Seeing what's in progress: the "noodling on" and "building" sections, keeping in mind many entries are one-line status updates.

Note that the site itself is marked WIP, with a link to an older version — so some links and sections may be incomplete or in flux.