About Francesco
I’m a Freelance Engineer with a strong background in robotics and applied research and I also hold a PhD with a Marie Curie fellowship. I help teams turn complex ML ideas into reliable, production-ready systems, working across the full pipeline, from data and modeling to deployment. My experience spans computer vision, deep learning, and real-world AI systems built to work outside the lab.
Italian
Native or bilingual
Spanish
Native or bilingual
English
Fluent
German
Conversational
Can work on-site
Forlì (up to 50km), Bologna (up to 50km)
Experience
- Self-employedComputer Vision AdvisorJanuary 2026 - June 2026 (5 months)Responsible of AI pipeline on medical visual information
- Lead the technical design of a self-supervised foundation model for gastrointestinal endoscopy: DINOv3 ViT-L/14 backbone with teacher–student self-distillation (DINO + iBOT losses), multi-crop strategy and k-NN evaluation monitoring; full PyTorch codebase with distributed (DDP) multi-GPU training.
- Integrate and curate large-scale multimodal health data: structured ~2.5 TB of raw clinical video into a reproducible training corpus of ~4.5M endoscopic frames across colonoscopy, gastroscopy and therapeutic procedures, addressing data quality, interoperability and AI amenability.
- Designed and implemented patient pseudonymisation (zero PII residual) and a provider-agnostic European data/compute setup (Hetzner storage; Nebius and CINECA Leonardo for GPU training) — hands-on with GDPR/EHDS-relevant constraints for health data.
- Built a benchmarking framework to assess robustness and performance against competing foundation models (EndoDINO, Endo-FM, EndoMamba) and to distinguish genuine advances from reproductions of known techniques; planned downstream evaluation for segmentation, pathology classification, severity scoring, phase recognition and lesion detection.
- University of Naples Federico IIDoctoral Research FellowDecember 2021 - February 2025 (3 years and 2 months)Naples, Metropolitan City of Naples, ItalyH2020 MSCA Project PERSEO• • Conducted analytical evaluation of interaction models and algorithms, assessing novelty, technical merit, and reproducibility; produced 7+ peer-reviewed papers• • Led investigations into personalization in robotics, coordinating interdisciplinary teams worldwide• • Supervised bachelor's and master's theses, mentoring students in structured research and scientific writing
- Roboception GmbHRobotics EngineerMay 2021 - November 2021 (6 months)Munich, Germany• • Designed and validated perception modules for industrial robot–vision systems (ROS, C++, Python)• • Enhanced grasping algorithms via performance assessment and optimization, improving reliability by 9%
Recommendations
Be the first to recommend Francesco
Help this freelancer shine by sharing your experience working together.
These freelancer profiles also match your criteria
Agatha Frydrych
Backend Java Software Engineer
4.7
(3)
2
Baptiste Duhen
Fullstack developer
4.6
(4)
5
Amed Hamou
Senior Lead Developer
4
(2)
7
Audrey Champion
Web developer
4.3
(3)
4
Education
- Ph.D.University of Naples Federico II2025Information and Communication Technology for Health
- M.Sc.University of Siena2018Computer and Automation Engineering - Robotics and Automation Track