Full-Stack Engineer: building products end to end. Two of my own, plus client work.
6+ years coding · Nigeria (UTC+1) · Open to remote roles worldwide
Six years of writing Python, Go, TypeScript, and Next.js, most of it spent on two questions: does this hold up once real people depend on it, and who fixes it at 2am when it doesn't. Quorel and VarsityLine are both still answering yes.
I built both of those solo, from schema to UI to the infrastructure keeping them online. The rest of what's here is client work and side projects.
Primary stack: Python · Go · TypeScript · Next.js Databases: PostgreSQL · MySQL
Quorel: live, versioned web data for AI agents
Gives your AI agent a live, versioned view of any website. Describe the fields you want in plain English, Quorel extracts them nightly and keeps every past version permanently queryable. Every dataset ships with an MCP server, so Claude or any MCP-compatible agent can query it, catch what changed since last week, and even fix bad rows itself, publishing the correction as a new, reversible version.
4 live datasets · full version history, nothing overwritten
View Site ↗ · Browse Live Datasets ↗ · GitHub
VarsityLine: verified Nigerian university admission data
Nigerian students figure out admission cut-offs from group chats and outdated PDFs. VarsityLine fixes that: verified cut-off marks, courses, and screening dates for every Nigerian university, cross-checked against each school's own admissions office, with a "last confirmed" timestamp on every entry. Search by university or course, filter by state and type, export as PDF, and subscribe for search-and-compare tools.
24 universities covered · verified cut-offs & courses · Paystack-powered alerts
View Site ↗ · GitHub
Uploading a 40-page PDF and getting a wall of text back isn't a summary, it's a chore. Skim lets you tell it exactly what you're looking for, then returns a clear, page-by-page summary in seconds, built for students and teachers who need the signal, not the whole document.
Real users · page-by-page summaries · targeted extraction, not a wall of text
View Site ↗ · GitHub
Store owners waste hours hand-writing SEO copy for every product listing page, and most of it still doesn't rank. Wildpalace plugs into Shopify and auto-generates SEO-optimized Product Listing Pages straight from keyword intent, turning a manual content chore into something that runs itself.
GitHub
More experimentation and older work on my GitHub if you want to dig deeper.
A real-time 3D ring viewer for jewelry e-commerce. Customers rotate, zoom, and inspect rings from every angle before buying, reducing returns and building confidence. 360° rotation with mouse and touch, real-time zoom/pan, multiple material & finish previews, embeddable in any store page.
Stack: Next.js · Three.js · TypeScript · Tailwind · WebGL
View Demo ↗ · GitHub
4 premium landing pages, including the Aqua Collection: a cool-toned jewelry store with a crisp aqua palette, airy modern layout, smooth hover transitions, and a streamlined checkout experience.
Stack: Next.js · TypeScript · Tailwind · Framer Motion
| Portfolio | var-raphael.vercel.app |
| samuelraphael925@gmail.com | |
| X | X |
Open to remote roles worldwide · Response time within 24 hours