Welcome to my own little plane of Oblivion! 🪐
My name is Mateus Begnini Melchiades and I am a Software Developer and Researcher in Machine Learning.
My main field of study is centered around Option discovery in Reinforcement Learning, but I also work on projects related to NLP and Neural Networks.
I work as a Software Developer at SAP Labs Latin America, where I develop backend services related to Data Science.
I also hold a bachelor's degree in Computer Science from the University of Vale do Rio dos Sinos (UNISINOS), with an outstanding student award from the Brazilian Computing Society (SBC).
On my spare time, I use my programming skills to create and contribute to open-source projects mostly related to Linux.
I spend most of this time improving Vanilla OS, which I act as Contributors Leader.
The programming languages I usually work with are (in order of familiarity):
- Python (including popular libraries like Pandas, Numpy, PyTorch, ...)
- Go
- C
- Lua
- Bash
- Rust
- Vala
My creations
- JABS.nvim: A minimal buffer switcher window for Neovim written in Lua.
- tree-sitter-vala: A tree-sitter implementation for the Vala programming language.
- Albius: An installer backend focused on immutable Linux distributions, but suitable for all distros (part of Vanilla OS)
Larger projects I contribute to
- Vanilla OS Contributors Leader: I develop and improve system applications like ABroot and Apx, as well as GNOME-related projects for better integration with the OS. My role in the project also involves managing incoming contributions and making sure our applications work as expected.
Research papers I worked at:
- Co-author of "FastIoT - A Compression Model for Displaying a Huge Volume of IoT data in Web Environments" (publication pending)
- Main author of "Anticipating faults by predicting non-linearity of environment variables with neural networks: a case study in semiconductor manufacturing" (accepted for presentation at LXAI @ ICML 2021)
- Co-author of "MoStress: a Sequence Model for Stress Classification" in 2022 International Joint Conference on Neural Networks (IJCNN), Padova, Italy, 2022.