About
I work on low-level systems software: operating systems, compilers, emulators, and developer tooling.
byteshiftlabs is my technical portfolio — both the projects I build from scratch and the open-source projects I contribute to. I'm always looking for new ones to get involved in.
Projects
Two kinds of work live here: projects I build for fun with AI — not only to learn their foundations in depth, but to test the limits of my AI tools — and open-source contributions to other people's projects I find professionally interesting.
Each project I build from scratch starts as rapid AI-assisted prototyping, then gets a proper engineering pass — architecture, implementation, testing, documentation — until it's something worth reading. Some projects I'm actively working on aren't mature enough to publish yet.
The full, current list of both — projects and contributions — lives on GitHub.
View projects on GitHubCertificates
Introduction to RISC-V (LFD110)
Getting Started with Rust (LFEL1002)
C++ Unit Testing: Google Test and Google Mock
Clean Code
Notes
Development Process
Projects here follow a two-phase process. The first is rapid AI-assisted prototyping. It is a fast way to get a working prototype, but not the end of the road.
The second phase is proper refinement: design decisions, clean implementation, testing, and documentation. The prototype becomes something technically honest and worth reading.
Workflow conventions and project self-configuration are handled through Rubric, a custom MCP server that enforces consistent standards across every byteshiftlabs repository.
Privacy
I won't work on any personal software projects that involve managing private data. I have no interest in dealing with data protection regulations there, and I'd rather keep it that way.
Platform
Projects here are developed and tested primarily on Ubuntu LTS. Compatibility with other distributions or operating systems is not guaranteed. Contributions targeting specific hardware or kernels (Zephyr, Linux) are built and tested per that project's own process instead.
Personal Tools
Some of the projects here are simply tools I built for myself to solve specific needs. Just practical solutions that worked for me.
Project Scope
Byteshiftlabs builds working implementations of systems chosen to learn their foundations in depth and to test the limits of AI-assisted development. The goal is a codebase that is correct, readable, and well-documented. Each repo's own README states its actual maturity and whether it's meant for production use.
AI Usage Policy
When contributing to someone else's project, I follow and adapt to that project's own AI usage policy, whatever it is.
My own projects are built with AI assistance. The architecture, interfaces, and error-handling strategy are decisions I make before any code gets written; I work through the implementation afterwards, reading it until I can explain it. That reading is the slow part, and I only get evenings and weekends for it — there will always be some piece of code I haven't touched yet.
Contact
If a repository is useful to your work or you want to collaborate on something you find interesting from a technical perspective, get in touch.