I’ve always wondered: what actually makes an email urgent? Is it the length? The words? The tone? So I built TriageAI - a model that classifies incoming emails into priority levels (Urgent, High, Medium, Low) to figure out what deserves attention right now.

I’ve always been curious about emergence. How does organized, purposeful behavior arise from simple rules? How does a colony of ants, with no central planner, manage to build complex structures and find food efficiently?

So I built OKIOS — named after the Greek οἶκος, meaning home or habitat. It’s a 3D artificial life simulation where creatures with neural-network brains learn to survive entirely on their own. No predefined roles. No hand-coded behaviors. Just pressure, time, and mutation.

I use PyTorch every day at work. It’s incredible - but I’ve always wondered: how does it actually work under the hood? How does backward() actually compute gradients through a neural network?

So I built my own minimal deep learning framework called Synap. It’s written in C++ for performance, with Python bindings via pybind11. No external ML libraries - just raw tensor operations and automatic differentiation from scratch.

I’ve always been curious about what happens inside game engines when they render a 3D model. You know that feeling when you use Unity or Unreal without understanding what’s actually happening on the GPU? That’s exactly why I decided to build my own 3D renderer from scratch using C++ and OpenGL.

No engines. No magic. Just raw graphics programming.

I’ve always been fascinated by programming languages. How does code actually become a running program? What happens when you type let x = 5; in a REPL?

So I built my own. Meet Bat - a tiny interpreted programming language written entirely in Go. It’s not useful for production, but it taught me how interpreters actually work.

Ever wanted to quickly understand what a GitHub repository looks like without cloning and exploring it yourself? I built a tool that does exactly that - paste any GitHub URL, and it analyzes the entire codebase, extracting complexity metrics, code structure, and language distribution.

The best part? It combines Go and Python in a way that showcases how to pick the right tool for each job.