vlune_

vly.me

19-year-old dev building high-performance AI systems.

// my favorite tools

The Stack

Rust

My language of choice. Zero-cost abstractions, fearless concurrency, and the satisfaction of making the borrow checker happy.

systems performance safe

HiP / AMD ROCm

GPGPU on AMD hardware. Writing kernels, optimizing memory access patterns, and pushing silicon to its limits.

gpu compute parallel

Machine Learning & AI

From training loops to inference optimization. Building ML systems that are fast, efficient, and actually useful.

training inference optimization

// current focus

In the Lab 🧪

In Progress

Optimizing ML Kernels

Writing custom GPU kernels for ML workloads — squeezing every last FLOP out of AMD hardware with HiP.

In Progress

Low-level Rust Exploration

Unsafe blocks, FFI boundaries, and custom allocators — going deep into Rust's lower layers.

Coming Soon

Open-source Projects

Tooling and libraries born from my experiments. Stay tuned — something is brewing. ☕

Coming Soon

Writing & Notes

Technical write-ups on GPU programming, Rust patterns, and things I learn along the way.

// about me

Hey there ✌️

Just a 19-year-old dev curious about how hardware and software talk to each other. When I'm not fighting the borrow checker, I'm exploring GPGPU with HiP.

I love the intersection of systems programming and machine learning — building things that are both fast and smart. There's something deeply satisfying about writing code that runs close to the metal.

// probably debugging something right now 🐛

// let's connect

Say Hello

Always happy to chat about Rust, GPU programming, AI, or anything tech. Feel free to reach out!

GitHub

@vlune

Discord

vlune

hello@vly.me