I turned my own portfolio into an IDE-style site, and what I would change
My portfolio used to look like every other developer portfolio. A large name at the top, a short line about building websites, a grid of project cards, and a contact form at the bottom. It was accurate and it was forgett
My portfolio used to look like every other developer portfolio. A large name at the top, a short line about building websites, a grid of project cards, and a contact form at the bottom. It was accurate and it was forgettable, including to me. I spend most of my working day inside a code editor, so last year I rebuilt the site to look like one. The live version runs at mehedimansurrafi.online, and it is the site I send with every proposal now. This post covers why I chose the layout, what using it has taught me, and what I would change if I started again.
Why an IDE layout
An editor window has a shape most developers read without thinking. A file tree on the left lists everything in the project. The main area shows one open file at a time. Tabs along the top show what else is open. A status bar at the bottom carries the small details.
My portfolio maps onto that shape with no forcing. The file tree became the menu: home, projects, each project as its own file, about, contact, resume. Each page reads like an open file, with the project pages written the way a readme is written, what the thing does first, then the stack, then the links. The status bar carries my location, my availability, and a contact link, so they stay visible on every page without a banner shouting about them.
The honest reason is simpler than the design argument. The layout says what I do before a visitor reads a word. A clinic owner may not care, and this site is not aimed at clinic owners. It is aimed at developers and at the clients who hire developers and have learned to read a GitHub profile. For that audience, an editor is a familiar room.
How visitors use it, against a normal portfolio
I have no analytics story to tell here, and I will not invent one. What I know comes from sharing the link and watching what people open, and from testing the site myself the way a visitor would.
Two differences stand out against my old layout. The first is that people treat the file tree as a table of contents and click it directly. On the old site, the path was scroll and hope: hero, about, projects somewhere below. Here the whole site is listed on the left from the first second, and a visitor with one question, what has he built, goes straight to the projects folder. The second difference is order. Visitors open the project files before the about page. On a normal portfolio the about section sits high and gets read early. Here the work reads first and the person reads later, which is the order I would choose for them anyway.
The cost is real too. A visitor who has never used an editor meets a busier first screen than a name and a grid. Some of the editor decoration, the line numbers and the tab shapes, is pure costume for them. I accept that cost because of who the site is for, but it is the first thing I would weigh if the audience changed.
What I would keep
The file tree menu stays. It shows the full site at a glance and it never moves, which a top navigation bar on a long page cannot do.
The project pages stay written like readme files. A fixed shape, problem, build, stack, live link, code link, means a visitor can read the second project faster than the first, because they already know where each fact sits.
The status bar stays. Contact details that are visible on every page, without a popup or a floating button, are easier to act on than details buried at the end of one long page. Every path through the site ends one click from the contact file, and that was a deliberate choice, not an accident of the layout.
The theme switch stays. Dark is the default because the editor idea asks for it, and light is one click away, built on one set of CSS variables so both themes are the same layout wearing different colours.
What I would change, and what I dropped
I started with too many files in the tree. Every small page got its own entry, and the list read like a directory dump. I merged the small pages and grouped the rest, and the tree got shorter and clearer. If I started again I would plan the tree at six to eight entries from day one.
I dropped the fake terminal. An early version had a console panel that typed out lines about my skills. It did nothing a paragraph could not do, it slowed the page, and it was the kind of clever touch that makes a visitor watch the site perform instead of reading it. Anything in an editor layout that does not help navigation or reading is costume, and costume has to earn its place. The terminal did not.
I would also move the resume earlier. It sat low in the tree at first, grouped with the about pages. Recruiters and clients ask for a resume more than they ask for anything else on the site, so it earned a place near the top of the tree. That change came from requests, not from taste.
The mobile problem
An editor is a wide screen tool. Shrink it to a phone and the layout fights you: the tree wants a third of the width, the tabs want a row, and the page gets what is left. My first mobile version squeezed all three in, and the text column was too narrow to read in comfort.
The fix was to stop treating the phone layout as a smaller editor. On a small screen the sidebar closes into a menu button, the tabs go, and the page becomes a single column that reads like a document. The file names survive as the menu list, so the idea is still there, but nothing competes with the text. I check the site at 390px width, the size I use for every client build, and the rule I follow is plain: if a page needs horizontal scrolling or pinch zoom at that width, the page is not done.
This is the part of the design I would warn anyone copying it about. The desktop layout is the fun half of the work. The phone layout is where the design either becomes a real site or stays a screenshot.
The template decision
Other developers asked how the site was built, and the easiest honest answer was to sell the design as a template. That version is FolioForge. Turning a live Laravel platform into a template meant cutting, not adding. The live site runs on a Laravel app with an admin side, member pages, and settings. A template buyer needs none of that. They need files they can open, edit, and upload anywhere.
So FolioForge is the static version. Plain HTML and CSS, no database, no admin panel, and a contact form that opens the visitor mail app, because a static site has no server to send mail. It keeps the parts that survived my own edits: the short file tree, the readme shaped project pages, the status bar, and the phone layout that drops the editor furniture. It drops the platform parts that only make sense on my own site.
FolioForge is this design finished as a template. It has 6 pages, dark and light themes, and a fictional sample persona you replace with your own details. It costs $19 on Gumroad: https://mehedirafi.gumroad.com/l/folioforge
I am Mehedi Mansur Rafi, a freelance Laravel developer (B.Sc. in Computer Science & Engineering). I build hospital, clinic, and small business websites, and I take website work through Fiverr: https://www.fiverr.com/mehedi__rafi. My full store is at https://mehedirafi.gumroad.com/ and my site is https://mehedimansurrafi.online
Originally published by Dev.to WebDev. Aggregated on AIWithGhost for educational purposes β full credit and traffic to the original publisher.

