← Back to Portfolio
Publish date: 2026-10-09

This Website: A NieR:Automata Portfolio

This website is my main project: the AbdulrahmanAlmyman repository itself. This page explains the idea behind the site, why it looks the way it does, and what I want it to do for me.
  • Live site: www.abdulrahmanalmyman.dev (also on GitHub Pages)
  • Source code: github.com/PYTHON01100100/AbdulrahmanAlmyman

The idea

A CV is a list of claims. A portfolio is supposed to be the evidence. My goal was a site where a recruiter, an engineer or a hiring manager can see, in a few minutes:
  • what I do: cloud infrastructure, Kubernetes, DevOps and AI/LLM platforms,
  • how I think: written case studies and blog posts that explain decisions, not only results,
  • what I have proven: certifications and real projects, each one with its own page.
So the site is built around content, not decoration. Every project and article is a plain Markdown file, so publishing a new one is a single file and a git push.

Why the NieR:Automata style?

Most developer portfolios look the same: a dark gradient, a hero section, a grid of cards. I wanted mine to be remembered.
NieR:Automata has one of the most distinctive interfaces in games: a flat beige-grey palette, thin outlines, hard offset shadows, a faint diagonal hatch and grid in the background, and menus that glitch when you select them. It is calm, a bit melancholic, and instantly recognisable. It also happens to suit a technical site, because it already looks like a terminal or a system log.
I translated that into the interface:
  • Palette and background: the parchment colours, the diagonal hatch and fine grid, and the dark bars at the screen edges.
  • Sections: headings written as code comments, like /* About */.
  • Boxes and rows: bordered panels with a hard offset shadow that invert their colours with a short glitch when you hover, like selecting a menu item.
  • Archives: the project list is a directory listing (ls -l /archives) with a blinking cursor at the end.
  • Data Logs: short, mission-log style updates, newest first.
  • InMemory Intel: my articles, shown as numbered memory entries.
  • Dark mode: the same palette inverted into warm charcoal with parchment text, saved in your browser and following your system setting by default.
I kept the effects restrained on the reading pages. A case study should be comfortable to read for ten minutes, so code blocks are a clear dark panel with a copy button, and the glow effects were removed from the article text.

How it is built

  • Next.js with static export, so the whole site is plain files that can sit on GitHub Pages with no server to run or pay for.
  • Markdown case studies with a small front-matter header (name, date, type, description). The type decides where an entry shows up: project entries go to the Archives, blog entries to InMemory Intel.
  • GitHub Actions builds and deploys the site on every push to main.
  • Tailwind CSS with a few CSS variables for the theme, which is what made the dark mode a small change instead of a rewrite.
  • Search-friendly metadata: a sitemap, robots.txt, structured data and bilingual keywords (English and Arabic), so people can find me whichever way they search for my name.

My own domain: Cloudflare DNS + GitHub Pages

A personal brand needs an address that belongs to you, so I bought the domain abdulrahmanalmyman.dev (the site lives at www.abdulrahmanalmyman.dev, and the bare domain redirects to it) and put its DNS on Cloudflare, while GitHub Pages hosts the site for free. Here are the steps if you want to do the same.

1. Put the domain on Cloudflare

  1. Buy the domain (from Cloudflare Registrar or any registrar).
  2. In Cloudflare, choose Add a domain and enter it, then pick the free plan.
  3. If you bought it somewhere else, copy the two nameservers Cloudflare gives you into your registrar's nameserver settings. DNS now runs from Cloudflare (this can take from minutes to a few hours).

2. Add the DNS records

In Cloudflare go to DNS → Records and add:
TypeNameContentProxy
A@185.199.108.153DNS only
A@185.199.109.153DNS only
A@185.199.110.153DNS only
A@185.199.111.153DNS only
CNAMEwwwpython01100100.github.ioDNS only
Those four IPs are GitHub Pages' addresses for an apex (root) domain, and the CNAME sends www to your GitHub Pages site. Check them against the GitHub docs before you use them, in case they change.
Keep the records on DNS only (grey cloud) at first. GitHub has to reach the domain directly to issue the HTTPS certificate.

3. Tell GitHub Pages about the domain

  1. Open the repository, then Settings → Pages.
  2. Under Custom domain, enter www.abdulrahmanalmyman.dev and save. GitHub checks the DNS records.
  3. When the check passes, tick Enforce HTTPS. The certificate can take a few minutes to appear.
The site is deployed with GitHub Actions, so you set the domain in these settings. You don't need a CNAME file in the repo.

4. Optional hardening

  • Verify the domain in your GitHub account's Pages settings (it gives you a TXT record to add in Cloudflare). This stops anyone else from taking over your domain name on GitHub Pages.
  • If you later switch the records to Proxied (orange cloud) for Cloudflare's CDN and protection, set Cloudflare SSL/TLS mode to Full (strict). Otherwise you can get redirect loops.
  • Redirect www to the root domain, or the other way round, so there is one canonical address.

5. Update the site itself

Once the domain works, the site no longer lives under /AbdulrahmanAlmyman/ but at the root. I updated the canonical URL, sitemap and Open Graph tags to use https://www.abdulrahmanalmyman.dev, so search engines index the custom domain.

A new version, and the old one

This project is the current portfolio. My first portfolio is now the old version, kept for reference only:
  • Old live site: python01100100.github.io/Abdulrahman_Almyman_Portfolio
  • Old source code: github.com/PYTHON01100100/Abdulrahman_Almyman_Portfolio
The old version was a static page that told people about me. The current one is a system that lets me keep publishing: a project archive, a blog, a log of updates, a theme, and a deployment pipeline. It is a portfolio I can grow for years, instead of one I have to rebuild.

The goal: a brand that is not limited to Saudi Arabia

I am proud of where I come from, and the Saudi tech scene gave me my start. But I want my name to mean something to engineers and employers anywhere, not only in the Kingdom. The site is designed with that in mind:
  • English first. Every case study is written in English, in the way international teams write: problem, approach, architecture, what went wrong.
  • Global credentials. Certifications from AWS and Alibaba Cloud, and write-ups on tools used all over the world: Kubernetes, vLLM, LiteLLM, and KEDA.
  • Local roots, visible. The Arabic keywords and name in the metadata keep me findable at home while the content speaks to everyone.
  • A recognisable identity. A memorable visual style makes the site, and so the name behind it, easier to recall than a standard template.

A CV that shows the work behind it

The second goal is to make my experience easy to verify. My time at SCCC (by stc) shaped much of my cloud, DevOps and AI platform skills, and I do not want that to be a single line on a CV. So the site connects each claim to evidence:
What a CV saysWhere the site shows it
Kubernetes and cloud-native skillsThe Kubernetes learning journey posts, the K3s homelab, KEDA and zero-downtime deployment case studies
Alibaba Cloud and AWS expertiseALB on ACK, serving LLMs on ACK, AWS cost-alert automation, the ACA and AWS certifications
AI and LLM engineeringImage captioning with AWS and Hugging Face, and the multi-model LiteLLM + vLLM platform on ACK
DevOps and automationEnd-to-end automated deployment on Azure, GitHub collaboration, and this site's own CI/CD
CommunicationThe articles themselves, written to teach, with diagrams and downloadable files
Instead of "experienced in Kubernetes", a reader can open a page and see how I set up a cluster, what broke, and how I fixed it.

What is next

  • More write-ups drawn from real platform work, with diagrams and reproducible manifests.
  • A sharper, evidence-based CV that links straight to these case studies.
  • Keeping the Data Logs alive, so the site shows what I am learning right now.
If you are hiring, collaborating, or just curious about the build, the code is open. Take a look at the repository, or get in touch through the contact links on the home page.