Francesco Mattioli
Lead Partnership Engineer at Ultralytics
Rome, Lazio, Italy
Summary
🤩
Rockstar🎓
Top SchoolFrancesco Mattioli is a Lead Partnership Engineer at Ultralytics in Rome with six years of experience bridging AI research and production deployments. Trained in cognitive neuroscience and advanced AI, he pairs a deep understanding of human cognition with hands-on ML engineering—contributing fixes and export optimizations to high-profile projects like Ultralytics YOLO to improve TFLite, OpenVINO and CoreML deployment. He focuses on making models reliably deployable across platforms while improving training, validation, and documentation to streamline partner integrations. Known for a pragmatic blend of curiosity and caution, he delivers systems that balance innovation, usability, and real-world robustness.
6 years of coding experience
Advanced School in Artificial Intelligence, Advanced School in Artificial Intelligence at Consiglio Nazionale delle Ricerche , Istituto di Scienze e Tecnologie della CognizioneLaurea triennale, Laurea triennale at Università degli Studi di PerugiaMaster's degree, Cognitive Neuroscience, Master's degree, Cognitive Neuroscience at Sapienza Università di Roma
Italian, English
Github Skills (13)
computer-vision10
pytorch10
machine-learning10
data-export10
exports10
deep-learning10
python10
exporter10
openvino9
tensorflow-lite9
tflite9
documentation8
onnx8
Programming languages (8)
Github contributions (5)
ultralytics/ultralytics
Apr 2024 - Apr 2025
Ultralytics YOLO11 🚀
Role in this project:
ML Engineer
ML EngineerContributions:215 reviews, 119 PRs, 301 pushes in 11 months
Contributions summary:Francesco primarily contributed to the Ultralytics YOLO repository by fixing bugs and improving various aspects of the codebase related to model export, training, and inference. They addressed issues with TFLite export, OpenVINO integration, and CoreML export, as well as fixing issues with the YOLOv10 model. Moreover, they implemented fixes to improve the training and validation, and improved the documentation of the codebase. These contributions suggest a focus on optimizing the models for different deployment platforms and improving the user experience.
pytorchdeep-learningyolov8object-detectiononnx
LABSS/PyPROTON-OC
Feb 2021 - Sep 2021
PROTON-OC is an agent-based model that explores the dynamics and processes that lead to recruitment into organized crime.
Contributions:1 release, 4 reviews, 119 commits in 7 months
model-basedagent-basedcrimeagentlead
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.