Low-level backend engineering in C, C++, and Rust — with Python when the problem calls for speed of writing rather than speed of execution.
"Computers are like air-conditioners. When you open Windows, it becomes useless."
— Linus TorvaldsI work at the layer most people would rather not look at: memory management, concurrency, and the real cost of every instruction. It isn't about reinventing the wheel — it's about knowing exactly how it turns.
I use C and C++ when I need full control over what the machine does. Rust when I want that same control with compile-time guarantees. And Python when the problem calls for speed of writing, not speed of execution.
Curious about operating systems, compilers, and anything that runs close to the metal.
A minimal kernel written in Rust, built from the ground up: it configures the hardware itself, loads and executes an .elf binary, and drops it straight into ring 3.
A compiled programming language written in C++, inspired by Rust's discipline but built for raw speed — safety and performance treated as the same goal, not a trade-off.
A chess engine written in C++, playing at roughly 2000 elo — search, evaluation, and move generation tuned by hand.
A Python API that takes a chess game and analyzes it move by move — rating each one (brilliant, good, best...) and generating natural-language commentary in Portuguese or English.
A minimal reimplementation of Git in Rust — the essentials, built from scratch, to understand what actually happens under every commit.
Neofetch for Windows, without the noise — no clutter, no irrelevant stats like battery percentage you could just glance down and check yourself.