Senior Software Engineer
A self-owned generative-AI SaaS that mounts any ring onto a real finger from a single photo. Upload a wrist shot, pick a ring from the catalogue — or drop in an image from any jeweller’s site — and an image-generation model returns a photoreal try-on within seconds. End to end: Laravel API, queued worker pipeline, PayPal billing, jeweller-facing dashboard.
A structured natural language for describing interfaces. One .eng file, compiled to React JSX, Laravel Blade, or plain HTML — your choice of LLM determines the output quality. Open source, distributed as an NPM package, designed to make UI scaffolding feel like writing a spec.
--- type: component name: FeatureCard props: - name: icon type: string - name: title type: string - name: description type: string --- Create a card, padding large, rounded, border gray: Style this with hover shadow, transition 200ms Create icon circle, bg primary, 48px, rounded: Icon {icon} size 24px, color primary Show {title} as heading, bold Show {description} as paragraph, muted
import React from 'react'; import { Star } from 'lucide-react'; export default function FeatureCard({ icon, title, description }) { return ( <div className="p-6 rounded-lg border border-gray-200 hover:shadow-md"> <div className="w-12 h-12 rounded-full bg-indigo-100 flex items-center mb-4"> <IconComponent size={24} /> </div> <h3 className="text-xl font-bold"> {title} </h3> <p className="text-gray-500"> {description} </p> </div> ); }
@props([ 'icon' => null, 'title' => '', 'description' => '', ]) <div class="p-6 rounded-lg border border-gray-200 hover:shadow-md"> <div class="w-12 h-12 rounded-full bg-indigo-100 flex items-center mb-4"> <x-icon :name="$icon" size="24" /> </div> <h3 class="text-xl font-bold"> {{ $title }} </h3> <p class="text-gray-500"> {{ $description }} </p> </div>
<!-- Generated from feature-card.eng --> <div class="p-6 rounded-lg border border-gray-200 hover:shadow-md"> <div class="w-12 h-12 rounded-full bg-indigo-100 flex items-center mb-4"> <svg width="24" height="24"> <use href="#star"/> </svg> </div> <h3 class="text-xl font-bold"> Make it easy </h3> <p class="text-gray-500"> A short description here. </p> </div>
Laravel
PHP
FastAPI
MySQL
Redis
React
JavaScript
TypeScript
Tailwind
Node.js
Docker
Nginx
Stripe
GitHub
NPM
AI tooling sits inside the day-to-day loop — architecture exploration, multi-file refactors, test scaffolding, doc generation, and pre-PR review. The result is measured velocity without losing the discipline of a senior code base.
Service layers, domain models, queues, observers, policies — Laravel applications designed to scale beyond the demo. Code reviewers can read it; ops can sleep through it.
OpenAI, DeepInfra, embeddings, retrieval pipelines, agentic flows. Practical AI that ships — not demos. From semantic search to portfolio assistants.
Stripe Connect, subscriptions, marketplace flows, payout dashboards, reporting. Money that arrives correctly, every time.
MySQL schema design, indexing, query plans, Redis caching, queues, cron orchestration. The unglamorous work that makes products feel fast.
React, JavaScript, Tailwind, Blade — front-end glue that ties admin dashboards, CRMs and portals into the backend behind them.
Docker, CI/CD pipelines, Linux servers, Nginx/Apache, code review culture, mentoring, technical documentation. End-to-end ownership — from architecture call to the green deploy pipeline at midnight.
Map domain boundaries, weigh tradeoffs, draft lightweight ADRs and data models. Decisions land before a line of production code is written.
Repo-wide context, surgical edits across services, observers and migrations. Big shifts land as small reviewable PRs, not weekend rewrites.
Feature tests, edge-case generation, inline PHPDoc, README and runbook updates — produced alongside the code instead of an afterthought.
Security, performance, dead code, naming, N+1s, missing indexes — AI-assisted audit on the diff before reviewers ever see it. Less noise in PRs, faster merges.
Academic background at the intersection of large language models, benchmarking, and the evaluation of AI-driven visualisation — the same questions that show up in production today.
