Alessandro Pietro Bardelli Email & Phone Number

Milan, Lombardy, Italy

Milan, Lombardy, Italy

Collaboration with different teams including PyTorch, AISpeech, MotionAlgorithms and Ads. - Contributor to TorchRL and tensordict OpenSource libraries with over 50 PRs and code reviews on Github.- Implementation of a GPT-based Reinforcement Learning from Human Feedback end-to-end training pipeline that exploits and showcases all TorchRL-HF functionalities.- Reduction by 82% of the onboarding time of a large public audio-visual dataset to the internal infrastructure.- LLaMa LLM onboarding to the internal speech framework and further training and inference optimisation by enabling quantization.- Exploration of spectral features and filtering effects on model performance for IMU-based gesture recognition.- Successfully landing of several C++ codebase optimisations generating over 815K$ in estimated cost savings.

Milan Area, Italy

Research and develop Machine Learning and Data Science models and solutions for private customers or European research projects in the Manufacturing and Space&Defence sectors. - Tech co-lead of Leonardo Digital Lab and mentor of three Data Scientist / Data Engineer.- Tech lead of H2020 EU-project ADMITTED: Advanced Data Methods for Improved Tiltrotor Test and Design.- Design and implementation of a data ingestion and processing pipeline framework.- Implementation of forecasting and tracking models to monitor flight hours, costs and failures of critical parts. - 2nd place and Special Mention prizes in the Leonardo's AIRtificial Intelligence Challenge by leading a five people team in the implementation of a ML-powered application to detect damaged helicopter blades based on acoustic feedback.

Parma

Research and develop key algorithms in Artificial Intelligence and Data Science on proprietary and public data by working closely with engineers and researchers from different fields.- Creation, for a major motor car company, of an artificial driver model offering human-like longitudinal vehicle control in order to simulate emission in real driving conditions. - Responsible for developing the core Machine Learning algorithms of a new medical platform designed to ease patient's rehabilitation in post-acute phase after stroke.

Milano

Design and implementation of mathematical models and software solutions using machine learning algorithms and Big Data technologies

Milano

Design and implementation of algorithms for fraud detection, retention strategy and competitive analysis in the Insurance and Banking sectors.

Master Of Science (M.S.), Mathematical Engineering. Major In Computational Science For Engineering And Artificial Intelligence, 101/110

Activities and Societies: Associazione ingegneri matematici (AIM), Coro Alpini PolitecnicoThesis: "A Machine Learning Approach For Mesh.

Bachelor Of Science (B.S.), Mathematical Engineering, 100/110

Activities and Societies: Associazione ingegneri matematici (AIM), Coro Alpini PolitecnicoThesis: "Automatic Racing Lines Generation.

Maturità Classica, High School Diploma In Classics, 98/100