|
Gabriel Tavernini
I am a Software Engineer at Google based in Zürich. I hold a Master's degree in Computer Science with distinction from ETH Zürich (GPA: 5.87/6) with a focus on Machine Intelligence and Software Systems, alongside a B.Sc. from Politecnico di Milano (110 cum Laude).
My interests span across Machine Learning, Computer Vision, and Software Systems. I have experience working at industry leaders including Databricks and Amazon, where I developed expertise in building scalable systems and ML infrastructure.
Email  / 
CV  / 
Github  / 
Linkedin
|
👨🏻💻
|
|
Experience
|
|
Working on Google Search.
|
|
Worked on the DevEx team using Go, Scala, Linux, and AWS. Improved remote development system start/stop times by 80%, saving 25 engineer hours daily. Received a return offer for 2026.
|
|
Worked in a team of engineers supporting scientists in the organization.
Expanded capabilities of an internal tool used for ML deployments on AWS for Amazon Transportation Services.
|
|
Publications
|
|
|
SuperFlex: Deformable Superquadrics for Point Cloud Decomposition
Gabriel Tavernini*,
Elisabetta Fedele*,
Tiago Novello,
Leonidas Guibas,
Marc Pollefeys,
Francis Engelmann
ECCV 2026
project page
/
arXiv
SuperFlex improves over SuperDec in terms of reconstruction quality, primitive expressivity (introduces bending and tapering), and robustness to occluded point clouds.
|
|
|
Do Generalisation Results Generalise?
Matteo Boglioni,
Andrea Sgobbi,
Gabriel Tavernini,
Francesco Rita,
Marius Mosbach,
Tiago Pimentel
ACL Findings 2026
arXiv
An analysis of LLM generalization performance and whether improvements across diverse benchmarks reflect robust generalization or are specific evaluation artifacts.
|
|
Education
|
|
GPA: 5.87/6 | Specialization: Machine Intelligence with Minor in Software Systems
|
|
GPA: 4.09/4.3
|
|
Grade: 110 cum Laude/110
|
|