Skip to content

Work

Selected projects, professional experience, and research background across medical AI and robotics.

My career has followed one thread: making AI and autonomous systems work outside the lab. I started in industrial and human-facing robotics, the hardest place to deploy a system, where it has to work safely around people and machines in conditions you do not control. The same discipline now goes into clinical AI, where a model has to pass validation and governance and hold up in a real hospital. Robotics taught me deployment. Medicine is where I apply it.

Projects

Medical AI

Foundation Model for Gastroenterology Imaging

Context
An IRCCS hospital needed a foundation model for endoscopic polyp detection, trained on more than 5 million gastrointestinal video frames.
Constraints
Strict GDPR compliance and NDA boundaries, cloud-GPU cost ceilings, and a need for reproducible experiments across model architectures.
What I did
Designed the SSL pretraining strategy (masked image modelling with DINOv3-style self-distillation), built the cloud data ingestion pipeline, and set up experiment tracking and fine-tuning protocols for classification, segmentation, and detection.
Outcome
A reproducible foundation-model pipeline with EHDS-ready data governance and a defined path to clinical validation and edge inference.
Self-Supervised LearningPyTorchMedical ImagingVision Transformers
Edge AI

Edge AI Occupancy Monitoring System

Context
A ceiling-mounted system to monitor occupancy and inter-person distance, driving lighting cues when distance violations occur.
Constraints
Real-time inference on an NVIDIA Jetson, accurate distance estimation from a single top-view camera, minimal occlusions at 3-6 m mounting height.
What I did
Built a compact detection-and-tracking pipeline, calibrated the camera and applied homography to project detections onto the floor plane, and published zone events over MQTT.
Outcome
A working edge deployment performing real-time people tracking, distance estimation, and MQTT-driven lighting on constrained hardware.
NVIDIA JetsonComputer VisionPyTorchYOLO
Robotics

Autonomous Navigation for an Industrial Cleaning Robot

Context
A commercial ride-on floor scrubber needed an autonomous indoor navigation stack for large facilities such as warehouses, retail spaces, and airports.
Constraints
Robustness to floor sheen, reflective obstacles and mixed lighting; human-safe behaviour; repeatable coverage routes; simple operator workflows.
What I did
Developed the ROS2/Nav2 navigation stack (SLAM, localization, and coverage path-planning) and validated it in simulation (CoppeliaSIM) and on the platform.
Outcome
Reliable autonomous navigation with repeatable coverage paths and smooth, human-safe motion in large indoor environments.
ROS2Nav2SLAMLidar
Robotics

Reliability Uplift in Industrial Bin-Picking

Context
Roboception's stereo-based perception products, the rc_visard smart sensor and the rc_cube edge-compute platform, power industrial pick-and-place. Grasping needed to be more reliable in real factory conditions.
Constraints
Challenging factory lighting and clutter, tight latency budgets, and deterministic production-ready behaviour for integration teams.
What I did
Designed and implemented core C++/Python software for rc_visard and rc_cube, improved grasp generation and validation in the rc_reason component, and tightened the sensing-to-planning data flow.
Outcome
Faster pick-and-place cycles and more stable grasp proposals under difficult lighting and clutter, with cleaner software boundaries for predictable field updates.
Computer VisionStereo VisionC++ROS2
View project →
Robotics

Non-Verbal Human-Robot Interaction with TIAGo

Context
A TU Wien research study on whether a robot's non-verbal behaviour, coordinated gaze and arm motion, makes its intentions legible to people during a collaborative task.
Constraints
Safe, collision-free motion around people; precise temporal coordination of head and arm; a controlled within-subject experimental design.
What I did
Built the interaction engine on the bi-manual TIAGo robot, using MoveIt motion planning, synchronized gaze control, and ROS state machines, then ran the user study.
Outcome
Coordinated gaze-and-arm behaviour measurably improved perceived intention clarity, engagement, and interaction fluency.
HRIMoveItROS2Motion Planning
Robotics

Emotion-Aware Mobile Robot: Motion & SLAM

Context
A Noosware research platform, an emotion-aware mobile robot, supporting a study on how affective cues shape social distance and navigation comfort.
Constraints
Consistent localization across trials; eliminating clock-drift and TF extrapolation errors; non-blocking audio so navigation stayed responsive.
What I did
Implemented ROS2 motion behaviours and a SLAM stack, disciplined time synchronization (NTP, launch ordering, TF validation), and a Bluetooth audio pipeline for affective cues.
Outcome
Robust, repeatable runs with consistent localization and no SLAM resets from time drift, with motion and audio cues that participants found legible and engaging.
ROS2SLAMPythonSensor Sync

Experience

Mar 2026 - Present
Member, Commission for Information Engineering · Ordine degli Ingegneri di Forlì-Cesena · Forlì, Italy
Jan 2026 - Present
Founder, Independent Practice · Forlì, Italy
Oct 2023 - Feb 2024
Visiting Researcher · TU Wien, Autonomous Systems Lab · Vienna, Austria
Jul 2023 - Aug 2023
Visiting Researcher · Noosware NV · Eindhoven, Netherlands
Jul 2021 - Dec 2021
Robotics Engineer · Roboception GmbH · Munich, Germany
Apr 2020 - Jun 2021
Autonomous Systems Developer · STtech GmbH · Munich, Germany
Apr 2019 - Mar 2020
Research Assistant · Technical University of Munich (MIRMI) · Munich, Germany
Aug 2018 - Dec 2023
Managing Partner · Edilrevi SRL · Forlì, Italy
Aug 2018 - Sep 2018
Internship · Disney Research Zurich · Zurich, Switzerland
Jan 2018 - Mar 2018
Technical Consultant · Daimler AG dealers · Mid-North, Italy

Education

2015-2018
M.Sc. Robotics & Automation · University of Siena · Siena, Italy
2011-2015
B.Sc. Management Engineering · University of Siena · Siena, Italy
Download full CV (PDF)

Research background

Peer-reviewed research in human-robot interaction and social robotics. This work on human-facing systems underpins how I approach clinical AI.

Full publication list on Google Scholar →

Talks

May 2026

Intervista su IA e sviluppi tecnologici

Story-Time, Canale Italia, Bologna

Una breve intervista su Intelligenza Artificiale e sviluppi tecnologici. Il modo in cui il mondo reale e il mondo virtuale si uniscono.

Mar 2024

Datemi un bit e solleverò il mondo

Istituto Salesiano Sacro Cuore, Naples

A talk on the implications of Robotics and AI in the future of our societies, presented to ~200 high-school students at the Teatro Auditorium Salesiano Salvo D'Acquisto.

Aug 2024

Too Close to You? A Study on Emotion-Adapted Proxemics Behaviours

33rd IEEE RO-MAN 2024, Pasadena, California

A study on emotion-adapted proxemics behaviours.

Mar 2024

A Rosbag Tool to Improve Dataset Reliability

ACM/IEEE HRI 2024, Boulder, Colorado

Late Breaking Report presentation on a tool for improving the reliability of ROS bag datasets in human-robot interaction research.

Aug 2023

Sweet Robot O'Mine: How a Cheerful Robot Boosts Users' Performance

IEEE RO-MAN 2023, Busan, South Korea

Presentation of the paper exploring how a robot's cheerful communication style impacts user task performance in a game scenario.

May 2019

The Role of Closed-Loop Hand Control in Handshaking Interactions

IEEE ICRA 2019, Montreal, Canada

Presentation of the paper on the role of closed-loop hand control in handshaking interactions.

Milestones

2026
May

Attended Exposanità, the international healthcare and assistance exhibition, held in Bologna, Italy.

Mar

Joined the Commission for Information and Electronics Engineering, Ordine degli Ingegneri della provincia di Forlì-Cesena.

Mar

Attended MECSPE 2026, the international trade fair for the manufacturing industry, held in Bologna, Italy.

Feb

Started a medical imaging consultancy project: foundation-model fine-tuning for endoscopy.

2025
Dec

Won 1st prize at the 2nd Startup Creation Lab, Università di Bologna, hosted at Laboratorio Aperto Forlì.

Sep

Launched updated portfolio showcasing recent projects, ML demos, and interests.

Mar

Joined the Ordine degli Ingegneri (Forlì-Cesena), Sector A, Information Engineering.

Have a project in mind?

Get in touch