Computer Vision
Engineer
With over 3 years of hands-on experience in software engineering and machine learning, I've distinguished myself by implementing cutting-edge solutions in the financial sector and pushing the boundaries of computer vision. Throughout my career, I've had the privilege to work with RGI Group S.p.A., enhancing data exchange services using REST APIs and WebServices. Moreover, at Politecnico Di Torino, I developed advanced object detection algorithms for the RoboMaster competition, leveraging the latest in Computer Vision architectures, and implemented a new hyperbolic space architecture based on Meta work 'Mask2Former'. These ventures have equipped me with key skills in Python, Java, JavaScript, and SQL, and hands-on expertise with libraries such as PyTorch, Scikit-Learn, and many more, including various optimization regarding GPU programming, applied to computer vision tasks. During my path in Reco 3.26 I developed skills and created projects on computer vision regarding anomaly detection, multi-tracking and reidentification, applied to real time CCTV.
🔸 Notable Achievements:
Pioneering improvements in Object Detection using ViT transformers
Surpassing Google's inference methods in incident detection through Computer Vision techniques
Developed real-time CCTV anomaly detection systems with 99%+ accuracy
Published technical translations bridging AI knowledge gaps in Italian market
Advertisement
Featured Projects
Real-time Object Detection System
End-to-end object detection pipeline using YOLOv8, deployed on edge devices for real-time inference with 95% accuracy. Implemented for RoboMaster competition with custom ViT transformers.
Financial Data Exchange Platform
Enhanced data exchange services for RGI Group S.p.A. using REST APIs and WebServices. Improved system performance by 40% and reduced latency significantly.
CCTV Anomaly Detection System
Real-time anomaly detection, multi-tracking and reidentification system for CCTV networks. Achieved 99%+ accuracy in incident detection, surpassing Google's methods.
Mask2Former Hyperbolic Architecture
Implemented new hyperbolic space architecture based on Meta's Mask2Former work. Advanced semantic segmentation with improved performance on complex scenes.
AI Consulting Solutions
Custom machine learning solutions for various industries. From predictive analytics to computer vision applications, delivering measurable business impact.
Publications & Books
Super Study Guide for Transformers in Natural Language Processing
Italian Translation & Technical Adaptation
Complete Italian translation and technical adaptation of an advanced guide on Transformers in Natural Language Processing. This comprehensive resource has helped hundreds of Italian developers and researchers understand cutting-edge AI architectures.
What You'll Learn:
- Master Transformer architectures from scratch
- Implement state-of-the-art NLP models
- Understand Italian technical AI terminology
- Apply transformers to real-world projects
Interactive AI Demos
Drop your image here or click to browse
• Supported formats: JPG, PNG, WebP
• Max file size: 10MB
• Model: YOLOv8 (80 COCO classes)
• Processing: Client-side inference
Technologies & Expertise
Deep Learning
Computer Vision
MLOps & Deployment
Programming & Tools
Let's Build Something Amazing
Ready to Transform Your Business with AI?
Whether you need computer vision solutions, AI consulting, or custom machine learning models, I'm here to help you achieve breakthrough results. Let's discuss your project and create something extraordinary together.