A personal hub built from the ground up. Primarily a portfolio, but also home to a blog, a self-hosted app dashboard, and more. Deceptively simple on the surface, the site is packed with advanced, custom-built features you won’t find in an off-the-shelf template. Every part of it was an opportunity to push boundaries, experiment with new technologies, and most importantly; learn.

Pages

Home / Portfolio — A curated summary of my experience, projects, and skills. The heart of the site.
Blog — Where I write about projects, tech, and whatever else catches my interest.
Projects — An in-depth look at my work, with descriptions, tech stacks, and links to homepages or repositories.
Dashboard — A unified view of all my self-hosted apps, with dynamic access control powered by Authelia’s OIDC component, displaying information only to those explicitly authorised to see it (via the OIDC group scope). This opens up the possibility for a zero-trust approach, which is what I have opted for with this specific site configuration.
User — Login page for Authelia accounts. Authentication is universal across the site. Log in once, and everything your group is entitled to opens up automatically.

The Background

A fully custom fluid dot grid, built from scratch on an HTML5 canvas in vanilla JavaScript. It responds to every form of input (mouse, touch, scroll, and keyboard), spawning directional waves and click ripples that propagate across the grid with physically-grounded momentum and decay.

Performance was a deliberate focus: typed Float32Array buffers, spatial culling, and a fast-path that skips all simulation when the grid is idle keep it smooth across everything from high-DPI desktops to low-power mobile. The canvas is oversized to the viewport diagonal and GPU-composited, so rotation and resizing never cause clipping or repaints.

It also supports a bioluminescent trail effect: a batched, intensity-bucketed^[1] glow that follows the cursor without impacting frame rate.

Bucketing is a form of quantisation, but the relevant Wikipedia page is primarily focused on signal processing. The math is the same, but the context is different. ↩︎