Fernando Castor: professore e ricercatore su Large Language Models, open source alternatives, IDE refactoring tools e pedagogia software engineering. Esplora AI tooling limits, PALE licensing e strategie didattiche efficaci.
5 articles from this blog
Argues that LLMs serve as a baseline for developer tools, not replacements, due to their general but non-specialized capabilities.
Argues that the term 'Open Source' is misleading for LLMs and proposes the new term 'PALE LLMs' (Publicly Available, Locally Executable).
A researcher's informal survey among software engineering academics reveals surprising insights into their personal use of automated refactoring tools in IDEs.
Summary of Nancy Leveson's ICSE 2020 keynote on software safety, hazards, and the distinction between reliability and safety in system design.
Explains how Apple's transition to its own silicon impacts Swift developers, covering universal apps, Rosetta 2, and running iOS apps on Mac.
