Doug Fritz
Principal Engineer · Google DeepMind
I'm a Principal Engineer on AI for Science at Google DeepMind, working on the systems, visualization and workflows behind scientific research.
If you're working on a hard problem in science, or building tools for the people who are, I'd like to hear from you.
Selected work
Current work first, then tools for seeing and reasoning about complex systems.
-
Antigravity for Science
Science in Google Antigravity: agents that work inside a researcher's own workspace, grounded in more than 30 life-science sources, including UniProt and the AlphaFold Database.
-
AI Studio Build
Tech lead for AI Studio's build experience, from the starter apps that launched with Gemini 2.0 to the Build tab at Google I/O 2025, where a prompt becomes a working web app.
-
Project Astra
Prototyping for Project Astra, a research prototype exploring a universal AI assistant.
-
Music AI Sandbox
Experimental music tools, developed together with the musicians using them.
-
TensorFlow graph visualization
Made the structure of deep-learning models legible: an interactive, collapsible view of a TensorFlow graph. Best Paper at IEEE VAST.
Journey
From art and computer science, through tools for seeing data, to AI for science.
The Divergent Foundation
The Art
BFA Fine Art
Oil painting, bronze sculpture, interactive artThe Science
BS Computer Science
Logic, algorithms, systems, robotsWorking with Golan Levin and the Processing community, where art, science and visualization came together.
Synthesis: code as a medium for visual expression and visualization.
Learning to See Data
A research internship in IBM Research's Visual Communication Lab, working on Many Eyes. Interface work in Yahoo's yHaus incubator, and early frontend engineering at Siri as a start-up before grad school.
At the MIT Media Lab's Fluid Interfaces Group, a master's degree focused on machine learning and data visualization: topic models and visual tools for exploring large collections of documents, and SpaceMarks, a zoomable workspace with a model that learned from how people arranged their information. Then out of the PhD and into industry, because Google was where the data and compute were.
The Canvas of the Web
On Google Creative Lab's Data Arts Team: the WebGL Globe, an open platform for visualizing geographic data in the browser, and team experiments that pushed what the web could do.
Making Machine Learning Visible
Bringing that craft to machine learning: early design and development of TensorBoard, and the TensorFlow graph visualization, which won Best Paper at IEEE VAST.
Tooling, model visualization and app UI for Google Clips, and web components and visualization in Colab.
The Interpretability Frontier
Building and leading DeepMind's visualization team: tools for interpretability, evaluation and research.
AIX: AI Experience
Leading the AI Experience team, prototyping how people work with AI, including Project Astra and Music AI Sandbox. Tech lead for AI Studio's build experience, from the starter apps to the Build tab launched at Google I/O 2025.
AI for Science
Now bringing all of it to science: the systems, visualization and workflows behind research.