What I write about
The things I keep coming back to are tensors, kernels, inference, tooling, and the uncomfortable gap between a clean paper idea and the software that has to run it. If a concept becomes clearer when you trace the shapes, memory, or execution path, it probably belongs here.
Background
- Machine Learning Engineer at Hugging Face.
- Previously worked on machine learning and AI research at Domyn.
- Studied Computer Engineering and Computer Science at the University of Pisa.
- Based in Italy.
Interests
Artificial intelligence is the center of gravity, but I also care a lot about algorithms, big data, data compression, Linux, and the practical craft of building with Python, Rust, and C++.