A bit about me
Hi, I am Fabio, a physicist by formation, working at the Fraunhofer Institute for Industrial Mathematics (ITWM).
After receiving my bachelors and Masters in physics in Brazil, I completed my PhD in a Theoretical and Computational Neuroscience program in Göttingen, Germany, where I started my research on designing bio-inspired computing systems and machine learning, which persists until today.
I am interested in all things related to bio-inspired computing, Ai and system or data modeling, from Neuroscience to Machine Learning.
skills
Science
Physics, dynamical systems,
bio-inspired computers and
Ai in general
Data Science
Azure cloud, data management,
ML solutions (DP-200, DP-100)
Organizational
ITIL4 framework,
Agile - Scrum,
Scientific writing
Coding
Fundamentals of: c++, c#, python, R
and Unity
Languages
Portuguese (native), English (fluent),
German (Basic)
Experience
Multi-cultural/language teams,
leading and advising team members
projects

Emergent Dynamical Intelligence (EDI)
We aim to develop code and code-free tools to generate Ai solutions for autonomous agents based on the dynamics between sensors and internal machine states

Analysis of Cycles in Asset-Network (ACAN)
We aim to stablish a set of technical procedures and models to facilitate and optimize the analysis of complex networks comprised of (financial) assets.

Heteroclinic Computing Systems
The aim of this project is to understand the requirements for a heteroclinic computer implementation.
.

Learning via controlled transitions into chaos
In this project we explore the potential of controllable phase transitions in layered neural networks to learn new patterns (solutions) during run time.