Diogo Vaz — a PhD student in the Distributed, Parallel and Secure Systems (DPSS) research area — has published a piece on his research work in the Portuguese newspaper Público (following Diogo Nunes’ recent piece in the same publication).

As Diogo recounts, his “work intends to revolutionize the process of developing fault-tolerant algorithms through the implementation of a new tool that allows to generate correct and efficient algorithms automatically.” Diogo is doing this by using generative AI and reinforcement learning techniques (the first capable of generating content, such as text and images, based on specific prompts and pieces of data, and the latter a model capable of learning through trial and error), with the plan of applying this newly developed tool to solving theoretical and practical problems, such as blockchain algorithms.

Diogo’s piece is published within the series “Conta-nos a tua ciência” (“Tell us about your science”), in which PhD students at the University of Lisbon are given an online space to share their research — as well as its importance and societal relevance — with Público readers.

You can read Diogo’s entire piece, in Portuguese, here.