I am a Senior AI Research Engineer and Project Leader at the Samsung AI Center in Warsaw, Poland, where I work in the Agentic Reasoning Lab. I lead research on fine-tuning, agentic reasoning, machine unlearning, safety, and alignment for large language and vision-language models.
I develop new research ideas, test how well they work on academic and real-world benchmarks, and turn them into papers, patents, and working prototypes. I am especially interested in making large multimodal models more capable, efficient, and reliable.
I received my Ph.D. in Computer Science from Sapienza University of Rome in 2025, where I was part of the Perception and Intelligence Lab led by Prof. Fabio Galasso. Before joining Samsung, I worked on research at Panasonic North America and UC Berkeley BAIR, and was a lead researcher and founding team member at ItalAI.
Outside of research, I enjoy spending time in nature, reading, photography, sports, and playing guitar.
Get in touch at paolomandica.p@gmail.com.
Updates
June 3, 2026 The code for GPart is now open source on SamsungLabs GitHub.
Research
Projects & Service
Open Source & Apps
Bare Habits — a minimalist habit tracker for iOS. Focus on what matters, one day at a time.reed — a tool for converting saved article pages into EPUBs, Markdown documents, and audiobooks.
Teaching
Teaching assistant for these Master's degree courses in Data Science at Sapienza University:
- Fundamentals of Data Science and Laboratory - Fall 2022
- Algorithmic Methods of Data Mining - Fall 2020
Academic Service
Reviewer for leading AI and computer vision conferences and journals:
- Neural Information Processing Systems (NeurIPS)
- European Conference on Computer Vision (ECCV)
- Empirical Methods in Natural Language Processing (EMNLP)
- International Conference on Image Analysis and Processing (ICIAP)
- IEEE Transactions on Circuits and Systems for Video Technology (TCSVT)
- IEEE Transactions on Image Processing (TIP)