At ICommunix, I refactored a large Angular application, reducing memory usage by 13% and improving performance by 20%. I also built real-time communication and queue management with AWS Lambda and SQS for a serverless omnichannel contact-center platform.
At Objex, I designed a Google Vertex AI data pipeline that prepares database records for model fine-tuning. I also maintained event-driven services on Google Cloud and implemented asynchronous processing that reduced API response time from 10 seconds to 4 seconds.
At StrikeReady, I built an API and service monitoring system that processed more than 150,000 daily records, with dashboards and automatic MS Teams alerts. I also led three engineers in an incremental framework migration and architected a real-time notification system supporting more than 2,000 concurrent connections.
I'm completing an MSc in Data Science at the University of Catania, where my coursework includes machine and deep learning, big data, and cloud computing. In a project on shortcut learning in image classifiers, I fine-tuned ResNet-50 and used Grad-CAM and masking interventions to study attention bias.

