Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

Machine Learning Engineer & Researcher specialized in Deep Learning and Generative AI Engineering. I architect custom AI solutions from the ground up, moving far beyond basic prompting. Currently developing proprietary Vision Transformer architectures and high-precision parsing algorithms for complex document understanding. Research Focus: Published author on Multi-Agent Systems and Automatic Prompt Optimization. I developed frameworks that explicitly outperformed previous SOTA solutions in efficiency and accuracy…Machine Learning Engineer & Researcher specialized in Deep Learning and Generative AI Engineering. I architect custom AI solutions from the ground up, moving far beyond basic prompting. Currently developing proprietary Vision Transformer architectures and high-precision parsing algorithms for complex document understanding. Research Focus: Published author on Multi-Agent Systems and Automatic Prompt Optimization. I developed frameworks that explicitly outperformed previous SOTA solutions in efficiency and accuracy

Antonio Sabbatella

Data Scientist, AI Engineer, AI Prompt Engineer, +2





Machine Learning Engineer & Researcher specialized in Deep Learning and Generative AI Engineering. I architect custom AI solutions from the ground up, moving far beyond basic prompting. Currently developing proprietary Vision Transformer architectures and high-precision parsing algorithms for complex document understanding. Research Focus: Published author on Multi-Agent Systems and Automatic Prompt Optimization. I developed frameworks that explicitly outperformed previous SOTA solutions in efficiency and accuracy…Machine Learning Engineer & Researcher specialized in Deep Learning and Generative AI Engineering. I architect custom AI solutions from the ground up, moving far beyond basic prompting. Currently developing proprietary Vision Transformer architectures and high-precision parsing algorithms for complex document understanding. Research Focus: Published author on Multi-Agent Systems and Automatic Prompt Optimization. I developed frameworks that explicitly outperformed previous SOTA solutions in efficiency and accuracy

Available to hire

Zurich, Switzerland

Machine Learning Engineer & Researcher specialized in Deep Learning and Generative AI Engineering.

I architect custom AI solutions from the ground up, moving far beyond basic prompting. Currently developing proprietary Vision Transformer architectures and high-precision parsing algorithms for complex document understanding.

Research Focus: Published author on Multi-Agent Systems and Automatic Prompt Optimization. I developed frameworks that explicitly outperformed previous SOTA solutions in efficiency and accuracy

Skills

Python

Data Science

Experience Level

Python

Expert

Data Science

Expert

Work Experience

Lecturer at Il Sole 24 Ore Business School

November 1, 2025 - Present

Delivered intensive training on AI for Finance, covering Neural Network architectures, Time Series forecasting, and supervised learning algorithms in Python.

Machine Learning Engineer at OneOfTech

November 1, 2025 - Present

Developing a production-grade algorithm for autonomous selection and routing of PDF parsers, optimizing data extraction accuracy. Engineering deep learning vision solutions based on Vision Transformers (ViT) fine-tuned for complex document classification.

Data Science Intern at KNIME

October 1, 2024 - February 1, 2025

Designed and implemented a Gen AI workflow integrating LLMs and Google APIs for automated content optimization with a human-in-the-loop validation layer.

Machine Learning Researcher at University of Milan Bicocca

July 1, 2023 - December 1, 2025

Pioneered a Bayesian optimization method for LLM prompt generation, reducing optimization time by 30% and outperforming state-of-the-art approaches.

Data Science Intern at OAKS Cloud

January 1, 2023 - August 1, 2023

Developed cross-lingual sentiment analysis models and deployed a multi-document RAG system using LlamaIndex and pgvector for semantic search.