Federico Berto

Hello world!

I am currently an AI Scientist at Radical Numerics, where I work on agentic AI for Recursive Improvement Systems Architectures and deep learning <-> kernel optimization, among other things.

I obtained my Ph.D. in the Industrial and Systems Engineering department at KAIST, where I was fortunate to be advised by Prof. Jinkyoo Park at the Systems Intelligence Laboratory (SILAB) .

My main research interests include decision-making and optimization in discrete and continuous spaces, including:

• Neural Combinatorial Optimization (NCO)
• Scientific Machine Learning (AI4Science)
• Agentic AI workflows for optimization and discovery

I love to work on open-source projects. I recently co-founded the AI4CO open research group and collaborate with DiffEqML.