@kaxokaze

MCNS (Modular Computational Neural System) is my attempt to build a computational “brain” of a fruit fly from real connectome data. I’m using Janelia FlyEM’s MaleCNS dataset — neuron skeletons, synapses, connectivity, neurotransmitters, and other biological data — as the foundation for a digital fly.

The biggest challenge is scale: hundreds of thousands of neuron morphology files and massive biological datasets have to be downloaded, validated, organized, and transformed without losing their biological meaning.

I’m proud that MCNS is being built as a real data-driven system rather than a simulated concept. The long-term goal is to turn the reconstructed neural architecture into something computationally explorable and eventually capable of modeling aspects of fly neural behavior.

The project is still actively under development, so the GitHub repository is the best way to follow and test what exists so far.

I made PandaTap, a website for a small NFC review-card project I’ve been working on. The idea is pretty simple: businesses can give customers a branded card that makes getting to their Google review page much easier.

I wanted the website to feel like an actual product website, not just a school/project website, so I spent a lot of time working on the design, layout, animations, responsive behavior, and the way the physical card is presented.

The hardest part was getting everything to feel consistent across the different pages while still keeping the site fast and easy to use. I’m most proud of the overall polish and the fact that it’s actually deployed and usable.

You can explore the homepage and the different sections/pages to see how the product works and how it could be used by salons, restaurants, clinics, and other local businesses.

https://pandatap.onrender.com/