SOFTWARE ENGINEER

MD. Raiyan Ur Rahman Jiyon

I build production-ready web applications and scalable backend systems using React, Next.js, and Node.js — with a strong focus on performance and maintainability.

👨‍💻

About Me

Software Engineer who builds products that drive business results

Hi, I'm Raiyan Jiyon (MD. Raiyan Ur Rahman Jiyon) — a Software Engineer with hands-on experience building full-stack web applications from concept to deployment. I specialize in React, Next.js, Node.js, and modern database technologies — delivering clean, scalable, and production-ready systems.

I've worked with startups and agencies to ship real products: from AI-powered platforms and e-learning systems to logistics dashboards and automation tools. I care about clean architecture, performance, and writing code that lasts. Whether contributing to a team or building systems end-to-end, I focus on engineering solutions that scale.

My Journey

Work Experience

Full-stack Developer

Current

CodeMyPixel · Contract · Narayanganj District, Dhaka, Bangladesh · Remote

Jun 2025 - Present · 4 mos

Led end-to-end development of robust MERN applications from concept to deployment, enhancing user experience and delivering powerful business tools.

Architected "Vocale VA Platform" with React 19 & Next.js 16 — engineered real-time voice chat interface using LiveKit integration with conversation history, speaker detection, and advanced state management.
Built enterprise-grade customer management dashboard with advanced search, filtering, tagging, and multi-organization support with role-based UI rendering and organization-scoped data visualization.
Developed responsive component library using shadcn/ui, Tailwind CSS v4, and Framer Motion animations with full i18n internationalization support and dark mode theming.
Engineered AI-powered content automation platform (TopFloorTrends) — reduced manual content processing by ~70% through automated ingestion → creation → distribution pipeline serving 500+ tasks.
Designed scalable Node.js + Express REST APIs with MongoDB, implementing rate limiting, caching, JWT authentication, and RBAC for secure multi-tenant architectures.
Implemented logistics platform (De Vrachtpatser) with optimized API response time <200ms and Google Places integration achieving 95% address accuracy.
Built form validation and state management using React Hook Form, Zod schema validation, and TanStack Query for efficient server state synchronization.

Frontend Developer

appmakers co · Contract · Khulna, Bangladesh · Remote

May 2025 - Jun 2025 · 2 mos

Built the Privatune e-learning platform UI from scratch with a performant, responsive, and scalable Next.js + TypeScript stack.

Implemented brand-new UI with 20+ reusable components and integrated complex backend APIs for seamless data flow.
Optimized performance — improved Lighthouse score from 62 to 90+ through code splitting, lazy loading, and image optimization.
Collaborated with product/design to launch the redesigned, production UI to 2,000+ active students.

My Work

Featured Projects

A selection of engineering projects and client solutions I've built. Each project solves real business problems.

Vocale VA Platform - AI-Powered Customer Interaction Dashboard

PRIVATE

A comprehensive AI-powered customer interaction platform featuring real-time voice communication, advanced customer management, and enterprise-grade dashboard capabilities for seamless business operations.

React 19

Next.js 16

TypeScript

Tailwind CSS v4

shadcn/ui

LiveKit

React Hook Form

Zod

TanStack Query

Framer Motion

Live Demo

AgriSmart - AI-Powered Farming Assistant

An AI-powered agricultural web platform designed to assist farmers with data-driven decisions, offering intelligent solutions for crop selection, disease detection, and management.

Next.js

React.js

TypeScript

Shadcn

Node.js

Express.js

MongoDB

Gemini API

Live Demo Source Code

Privatune - Educational Learning Platform

PRIVATE

A comprehensive educational platform where students can enroll in admission-related courses for Bangladeshi private universities, watch videos, attend exams, and purchase educational materials.

React.js

Tailwind CSS

Node.js

Express.js

MongoDB

Firebase Authentication

Stripe API

Live Demo

Digital Electronics Hub

A full-stack e-commerce platform for digital electronics enthusiasts, allowing users to explore, save, and purchase products with integrated AI chatbot assistance.

Next.js

Tailwind CSS

Node.js

Express.js

MongoDB

JWT

Gemini API

Recharts

Live Demo Source Code

TopFloorTrends - AI-Powered Content Automation

PRIVATE

A full-stack AI automation platform to empower content creators, handling the entire workflow from content ingestion to distribution.

React.js

Node.js

Express.js

MongoDB

Tailwind CSS

TypeScript

AI APIs

Live Demo

Job Box - Modern Job Portal Platform

A platform that connects job seekers, recruiters, and administrators, enabling new job management, applications, and recruitment processes with role-based authentication.

React.js

React Router

Tailwind CSS

Node.js

Express.js

MongoDB

Firebase Authentication

Live Demo Source Code

RentRental - A Car Rental Platform

Your ultimate car rental platform to explore, manage, and book your preferred vehicles seamlessly.

React.js

React Router

Tailwind CSS

Node.js

Express.js

MongoDB

Firebase Authentication

Live Demo Source Code

MovieEra

Your ultimate movie portal to explore, manage, and favorite your preferred films.

React.js

React Router

Tailwind CSS

Node.js

Express.js

MongoDB

Firebase Authentication

Context API

Live Demo Source Code

CareSync AI - AI-Powered Healthcare Assistant

An intelligent healthcare platform that automates appointment scheduling, prescription reminders, report summarization, and follow-ups for patients and doctors.

React.js

Node.js

Express.js

MongoDB

TypeScript

Tailwind CSS

AI APIs

Twilio

Live Demo Source Code

Federated Learning for Healthcare ML

A comprehensive research framework for federated learning applied to healthcare machine learning, featuring privacy-preserving training, Byzantine-resistant aggregation, and clinical safety optimization achieving 87.04% recall on diabetes prediction.

Python

scikit-learn

TensorFlow

Federated Learning

Differential Privacy

NumPy

Pandas

Matplotlib

Source Code

Fed-EdgeCrop: Federated Learning for Plant Disease Detection on Edge Devices

A comprehensive system combining federated learning and knowledge distillation for privacy-preserving plant disease detection on edge devices, achieving 1.76 MB quantized model with 14.5 ms latency (69 FPS) on Samsung Galaxy A31.

PyTorch