Delio Vicini is a Senior Research Scientist based in Zurich with 12 years of experience specializing in physically-based differentiable rendering and inverse light-transport problems. He has driven improvements in both accuracy and efficiency of renderers, contributing substantively to the widely used Mitsuba 2 and Mitsuba 3 engines—especially volumetric path tracing, numerical robustness, and core pipeline features. At Google he bridges rigorous academic research with production-grade systems, building on a PhD from EPFL and internships at Facebook Reality Labs and Disney that focused on novel volume models and denoising. His work uniquely combines principled light-transport differentiation with practical speedups—such as leveraging generative models and gradient-domain techniques—to make inverse rendering tractable.
12 years of coding experience
2 years of employment as a software developer
Bachelor of Science (BSc) Computer Science (mathematics minor), Bachelor of Science (BSc) Computer Science (mathematics minor) at University of Bern
Master of Science (MSc) Computer Science, Master of Science (MSc) Computer Science at ETH Zürich
Mitsuba 2: A Retargetable Forward and Inverse Renderer
Role in this project:
Back-end Developer
Contributions:124 commits, 5 PRs, 84 pushes in 2 years 2 months
Contributions summary:Delio primarily contributed to the development of the Mitsuba 2 rendering engine. Their commits focused on improving the core object handling with a ref<T> check, including enhancements to various parts of the rendering pipeline such as the texture resolution function, path tracer, and intersection routines. They also focused on adding new rendering examples and improving the documentation.
Mitsuba 3: A Retargetable Forward and Inverse Renderer
Role in this project:
Back-end Developer
Contributions:12 reviews, 89 commits, 43 PRs in 1 year 10 months
Contributions summary:Delio contributed significantly to the `mitsuba3` rendering engine, primarily focusing on core features within the volumetric path tracer. Their commits addressed numerical issues, added support for disabling next event estimation (NEE), and incorporated changes for the Russian roulette path termination criterion. Further contributions included implementing volumetric rendering techniques and performance improvements and fixing minor issues in existing source code.
mitsubarendererinverseforward
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.