In 2017, I joined the University of Madeira as an Invited Assistant Professor where I’m currently teaching programming, IoT and networking courses. (see more)
My first research experience and therefore contact with IT was in 2010 during my MSc in Computer Engineering at ISCTE University Institute of Lisbon where I managed to publish several articles in the fields of Serious Gaming and Machine Learning. I pursued the topic of Machine Learning and submitted my master thesis with the title of Hierarchical Reinforcement Learning: Learning Sub-goals and State-abstraction. After obtaining my MSc degree, I was hired to the position of Invited Assistant Professor at ISCTE University Institute of Lisbon where I taught Introduction to Programming and Object-Oriented Programming courses over a time period of two years.
By the end of 2010 I started working at Microsoft Portugal as a Software Development Engineer and was responsible for developing Facebook applications (client-side) for crowdsourcing and data-speech collection, and was also part of an engineering team responsible for creating a web portal for accessing, debugging, processing and labeling live logs related with speech end-user experiences in Microsoft products. In 2013 my manager probed my interest to pursue a PhD Studentship in Industry at Microsoft developing machine learning techniques for recognizing and predicting human activities from RGB-D data using Kinect. Since the topic was fully aligned with my research interests, I accepted the challenge and managed to complete my Phd in Computer Science with an specialization in Machine Learning and Human Activity Recognition, always with a connection to IT.
Shortly after submitting my Phd dissertation in 2017 I embraced a new professional challenge and joined Axians from Vinci Energies as a Lead Data Scientist and until today I am managing a team developing multiple projects related to Machine Learning and Computer Vision. The projects range from developing a data analytics platform for managing car-parking facilities. Developing and applying machine learning techniques for image segmentation and classification of corrosion in metal structures. Implementing techniques for human detection and tracking in thermal and RGB-D images.
Also, in 2017, I joined the University of Madeira as an Invited Assistant Professor where I’m currently teaching programming, IoT and networking courses. In hindsight, the decision to pursue and obtain a PhD changed the way that I approach challenges in my professional and personal life, imbuing me with tools that will help me to overcome those challenges in years to come. It has also opened doors to opportunities that otherwise would not be presented.
I completed my PhD program in 2016, and left IT to start working as a Postdoc in the University of Pittsburgh, Pittsburgh, USA. (see more)
I joined newly created DaRTES lab in IT Porto as a researcher in 2010, after completing my Master degree in the University of Aveiro. Shortly after joining the group, I started my PhD in the University of Porto under the supervision of Luis Almeida and Pedro Lima. In the following years, I developed my thesis work in wireless communications and relative localisation for mobile agents, with focus in ad-hoc teams of mobile robots. That period was a great learning experience as I also had the opportunity to collaborate in several research projects being developed within the group, ranging from distributed control systems to large scale reliable video streaming. I completed my PhD program in 2016, and left IT to start working as a Postdoc in the University of Pittsburgh, Pittsburgh, USA. In my current position my research focus changed to a different topic: reproducible computationally-driven science. Recent awareness of a reproducibility crisis in science has made this a topic of increasing importance within every scientific community. In fact, governments, institutions, publishers, and communities themselves are pushing for increased transparency and reusability of research artefacts, and for reproducible published results. As part of that work, I'm working within my research group in the development of Occam, available in https://occam.cs.pitt.edu. Occam is an experiment management system with focus on preserving all the data and metadata used in an experiment, from software and datasets, to configurable parameters essential to replicate previous conclusions. Notably, it enables the reproducibility of experimental results by preserving their provenance, thus allowing to trace each result to the experiment that was executed to generate it. I have also been involved in several Artefact Evaluation committees, a process that is being increasingly adopted by many major conferences, and that intended to help testing and verifying the software artefacts that were used to produce paper's main results. Recently, I have been teaching a computer science course on computer organization and assembly language as a part-time instructor in the University of Pittsburgh.