Stefan Milosevic
AI/ML Research Engineer working in precision medicine, high-performance computing, digital twins, and foundational machine learning modeling.
Stefan is an AI/ML Research Engineer and Computational Biologist with 5+ years of experience across software engineering, machine learning systems, and translational oncology research. He builds end-to-end platforms that integrate single-cell, genomic, imaging, and clinical data into clinically actionable decision-support tools.
His research at the University of Cambridge (MPhil in Advanced Computer Science) was conducted under Prof. Pietro Liò and Prof. Dr. Sarah Teichmann at the Wellcome Sanger Institute, centered on graph neural networks for network-driven multiomic modeling in genomics. A Chevening & Cambridge Trust Scholar and recognized Cambridge Trust Impact Leader. Stefan's current focus is on digital twin frameworks for CNS oncology, modeling tumor evolution , and therapy response .
2024 – Present Member of Technical Staff, Health Team
Stealth Biotech
A computational biology lab building patient-specific digital twins for oncology
2024 – Present AI (Life Sciences) Lead, Advisor
BIO4 Campus
Organizing Committee · 6th Belgrade Bioinformatics Conference (BelBi) 2026
2024 – 2025 Visiting Postgraduate Researcher
University of Cambridge
2023 – 2024 Graduate Researcher
Cambridge Centre for AI in Medicine
2022 – 2023 Software Engineer & Data Scientist
BlueGrid.io
2021 – 2022 Technical Team Intern
Jozef Stefan Institute & Institute of Oncology Ljubljana
IT/Developer at REKONIO – Regional Congress of Internal Oncology, Ljubljana
2021 – 2022 Data Scientist Intern
Microsoft
PhD in Artificial Intelligence
University of Belgrade · Accepted 2024 (full-ride) · Ranked 1st · Not pursued
Thesis: Multi-Modal Graph Learning for Neuro-Oncology: Digital Twin Modeling of CNS Tumor Evolution and Therapy Response
MPhil in Advanced Computer Science
University of Cambridge · Collaboration with Wellcome Sanger Institute
Thesis: scMultiGraph: single-cell Multiomic modelling with Message Passing Graph Neural Networks
CULTC Mens Blues Tennis · Churchill College
BSc + MSc in Computer Science & Data Science (Computational Biology)
Singidunum University
Thesis: Improving Melanoma Detection through Enhanced Data Exploration, Image Augmentation and Deep Neural Networks
Google Developer Group on Campus · University of Belgrade
Cambridge Impact Leader
Cambridge Commonwealth, European & International Trust, University of Cambridge
National PhD Full-Ride Scholarship Offer
Republic of Serbia, University of Belgrade
Chevening Scholarship
Foreign, Commonwealth & Development Office
Cambridge Trust Scholarship
University of Cambridge
Science and Research Full-Ride Scholarship
Undergraduate Studies
AI in Oncology, chapter in Artificial Intelligence and Headache Disorders (Springer Nature, Headache Series)
Springer Nature · Forthcoming 2026 · Co-authored book chapter
The COVID-19 Images Classification by MobileNetV3 and Enhanced Sine Cosine Metaheuristics
Optimizing Convolutional Neural Network by Hybridized Elephant Herding Optimization Algorithm for MRI Classification of Glioma Brain Tumour Grade
Feed-forward Neural Network Training by Hybrid Bat Algorithm
Multi-layer Perceptron Training by Genetic Algorithms
RTS National Television Serbia
AI and Biotechnology in the Fight Against Cancer (Stefan Milosevic) | RTS Zdravstveni dnevnik
Panellist & Workshop Lecturer Belgrade Bioinformatics Conference (BelBi 2026)
Creator & Hackathon Judge Belgrade Bioinformatics Conference (BelBi 2026)
Workshop Lecturer Google Developer Group on Campus · Google Nexus: AI & ML in Practice · University of Belgrade
Hackathon Judge Google Developer Group on Campus · Google Nexus: AI & ML in Practice · University of Belgrade
Keynote ADOS · 2nd Open Balkan Digestive Oncology Expert Meeting
Keynote DSC Europe · DigiHealth Conference
Speaker University of Oxford · Reuben College
Panellist Serbian Medical Society · Oncology Section