David Stutz is a research scientist at Google DeepMind with 13 years of experience applying rigorous machine learning research to real-world generative-AI safety and provenance problems. He helped develop and ship SynthID (image/video watermarking deployed in Vertex AI), led Med-Gemini evaluation work and contributed safety evaluations for Bard/Gemini, while filing patents and publishing on conformal prediction under ambiguous ground truth. He blends theoretical contributions—developing conformal methods with finite-sample guarantees—with hands-on engineering, leading cross-team projects, running large-scale experiments, and mentoring researchers and students. His work spans adversarial robustness, uncertainty estimation, and practical tooling, with much of his code open-sourced on GitHub. Early in his career he also built and sold web applications and contributed front-end improvements to a Bootstrap multiselect plugin, illustrating an uncommon breadth from research theory to product delivery.
JQuery multiselect plugin based on Twitter Bootstrap.
Role in this project:
Front-end Developer
Contributions:9 releases, 1 review, 475 commits in 9 years 10 months
Contributions summary:David contributed to the project by adding button customization options and updating other options. They implemented button customization using CSS classes. The user's work involved modifications to the project's index.html and js/bootstrap-multiselect.js files, implementing HTML-based options and templates. The user's work also involved updating styles of the plugin.
Contributions:8 commits, 6 pushes, 12 comments in 4 years 4 months
pytorchextendeddeep-learningevaluationbenchmark
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.