Valentina Salvatelli
Research Lead at Ellison Institute of Technology Oxford
London, England, United Kingdom
Summary
👤
Senior🎓
Top SchoolValentina Salvatelli is a research leader with 11 years of R&D experience at the intersection of AI, biomedical imaging, and astrophysics, currently leading work on multi-modal and generative AI for health. She has driven teams from small research groups to global organizations, shipping production ML systems at IQVIA and contributing engineering improvements to Microsoft's InnerEye medical imaging library. Her academic background—a Ph.D. in Astrophysics with expertise in Bayesian statistics—underpins research published in major ML and physics venues and collaborations with NASA’s Frontier Development Lab on synthetic data and compression for space instruments. Recognized among the 50 most inspiring Italian women in tech, she blends rigorous statistical thinking with practical deployment experience across cloud and healthcare settings. Notably, she bridges domain science and engineering, routinely translating complex imaging and longitudinal data problems into scalable, production-ready solutions.
11 years of coding experience
9 years of employment as a software developer
High School, High School at Scientific High School P.RuffiniDoctor of Philosophy (Ph.D.) Astronomy and Astrophysics, Doctor of Philosophy (Ph.D.) Astronomy and Astrophysics at Sapienza Università di Roma
Italian, English, French
Github Skills (8)
medical-imaging10
pytorch10
azure10
deep-learning10
microsoft-azure10
python9
computer-vision9
mlops8
Programming languages (2)
Github contributions (5)
microsoft/InnerEye-DeepLearning
Jul 2021 - Feb 2022
Medical Imaging Deep Learning library to train and deploy 3D segmentation models on Azure Machine Learning
Role in this project:
ML Engineer
ML EngineerContributions:118 reviews, 14 commits, 13 PRs in 6 months
Contributions summary:Valentina primarily contributed to the development and improvement of the medical imaging deep learning library. Their work involved updating documentation for the environment and hello_world model, as well as generalizing SSL functionality for use with different datasets. They also added a new DeepMIL panda container and enabled DeepSMILE on large encoded datasets by encoding data in chunks and loading cached encoded datasets in CPU. Finally they fixed PandaInnereyeSSLMIL and updated hi-ml to the latest version.
deep-learningmachine-learningazure-machine-learningdeep-learning-libraryimaging
vale-salvatelli/sdo-autocal_pub
Jan 2021 - May 2021
Code for the auto-calibration project on SDOML data
Contributions:2 releases, 6 commits, 1 PR in 4 months
calibrationauto-calibration
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.