A collection of mistakes.
Hi, I’m Russell. I taught mathematics for a long time, wandered into computers and networks, and lately spend most of my evenings on machine learning and a home lab that is larger than it needs to be. This is where the projects, notes, stories, and lessons end up. If you’re here about work, the short version is a scroll away and the résumé is one click.
The short version.
For anyone here in a hiring capacity: a few verifiable things, stated plainly. The résumé page has the rest; everything else on this site is context.
Part of Team Reki, ARC Prize 2026
Our team placed second in ARC-AGI-3 Milestone Prize #1 — a reasoning benchmark built to resist memorisation. I contributed experiments run on the home lab’s GPUs.
Some Kaggle placings
18th of 2,963 in the 2026 NeuroGolf Championship; 78th of 4,729 in Orbit Wars; 123rd of 4,182 in NVIDIA’s Nemotron Model Reasoning Challenge; 309th of 4,186 in AI Agent Security: Multi-Step Tool Attacks.
Two papers, a while ago
Written as an undergraduate, still cited occasionally. Kirk & Wang, “Largest Number of Subtrees of Trees with a Given Maximum Degree,” SIAM Journal on Discrete Mathematics 22(3), 2008 · Kirk & Wang, “The Hyper-Wiener Index of Trees,” Congressus Numerantium 187, 2007.
Dubious wisdom / selected at random
“Whatever you do will be insignificant, but it is very important that you do it.”
Things worth keeping.
Not a polished chronology. Just selected work, experiments, visual pieces, and documentation that say more than a list of job titles would.
Made things
Software, automations, experiments, and the small tools that exist because doing something twice felt excessive.
Made-up things
Short stories, visual work, and other random things that don’t fit on a résumé.
Running things
Homelab, networking, services, segmentation, and abstract diagrams that wisely leave out the useful reconnaissance details.
Useful doors.
The public parts of the namespace. Some are finished. Some are merely optimistic DNS records.
A few things I share.
What I’m working on right now, and the hardware and software behind it.
Russell Kirk.
B.A. in Mathematics (University of South Florida, 2010), with two peer-reviewed graph-theory papers written as an undergraduate. Seven years teaching IB Higher Level Mathematics, AP Calculus and AP Statistics in Wuhan, China; then network and computer-lab support at Florida colleges; then graduate computer-science coursework through Georgia Tech’s OMSCS — Python, computer networks, cryptography.
Lately the work is machine learning — reasoning benchmarks, model evaluation, AI agents — done in competitions and on my own hardware, learning as I go. I like systems that explain themselves, tools that remove repetitive work, and projects with a little personality. Based in Lakeland, Florida; open to interesting work.
- Machine learningEvaluation & reasoning
- PythonTools that last
- GPU computingSLURM · CUDA · DGX
- NetworksSegmented, documented
- MathematicsThe through-line
Say hello, eventually.
The address above is rendered as an image. The buttons decode it locally in your browser; the plain address is not printed in the page markup.