Hello, I'm

Md Nahidul
Islam

Software Engineer specializing in MERN/PERN stack. Building scalable systems with System Design principles—Load Balancing, caching, and message queues. Passionate about AI & Data Science, aiming to integrate ML into robust data-driven solution.

Narayanganj, Bangladesh

About

Who I am

I am a Software Engineer, specializing in building scalable, production-ready applications with the MERN stack and PostgreSQL. I have hands-on experience implementing System Design patterns, including Redis caching, BullMQ for background tasks, and real-time WebSocket communication.

Alongside web development, I am a researcher in Computer Vision and Explainable AI (XAI). My undergraduate thesis focuses on skin lesion classification using Swin Transformers, achieving 98.3% accuracy on the HAM10000 dataset.

I am a fast learner with an engineering mindset, dedicated to bridging intelligent, data-driven ML solutions with robust, user-centric web architectures.

Moving forward, I am dedicated to evolving into the AI & Data Science domain, aiming to bridge the gap between intelligent ML models and robust, user-centric web systems to build the next generation of data-driven applications.

Download CV

3.74

CGPA

10+

Projects

627+

GitHub Contributions

Feb 2026

Graduated

LocationNarayanganj, Dhaka, Bangladesh
Emailnahidulislam019045@gmail.com
LanguagesBengali (Native), English (Fluent)
InterestsNew Technologies, AI, Machine Learning, Data Science, Computer Vision, NLP, Deep Learning, System Design

Professional

Work Experience

Jr. Full Stack Developer

Beuptech Agency

Feb 2026 - PresentFull-time

📍 On-site

Designing and implementing scalable, high-performance full-stack solutions with a strong focus on robust system design principles. Architecting modular, highly available backend infrastructures, containerizing services with Docker, and deploying them efficiently on AWS. Optimizing data models, database queries, and API endpoints while implementing caching and rate-limiting strategies to handle high-concurrency traffic seamlessly.

AR VR Developer

Battery Low Interactive Ltd.

Nov 2025 - Dec 2025Internship

📍 Dhaka, Bangladesh · On-site

Collaborated on designing and developing immersive Augmented Reality (AR) and Virtual Reality (VR) applications. Contributed to building interactive environments, optimizing real-time rendering, and implementing core application logic to deliver high-quality user experiences.

Expertise

Skills

Front-End

JavaScript
TypeScript
React
Next.js
HTML5
CSS3
Tailwind CSS
Material UI
Bootstrap
Redux Toolkit
RTK Query
TanStack Query
Vite

Back-End

Node.js
Express.js
MongoDB
Mongoose
PostgreSQL
Prisma
Redis
Socket.io
JWT
Zod
REST API
BullMQ

Cloud & DevOps

AWS
Docker
Git
GitHub
Firebase
Vercel
Postman
Stripe
Cloudinary
VS Code

AI & Machine Learning

Python
PyTorch
NumPy
Pandas
scikit-learn
Matplotlib
Deep Learning
Computer Vision
Swin Transformer
Explainable AI

Work

Projects

Elite Property

Full-Stack Property Management System with role-based access for Admins, Agents, and Users.

  • Developed a role-based access control (RBAC) system for Admins, Agents, and Users.
  • Integrated Stripe API for secure payment processing and transaction management.
ReactNode.jsExpressMongoDBFirebaseTailwind CSSStripeTanStack Query

Choose Your Own Car

Production-ready Car Marketplace REST API with real-time chat, payment gateway, and advanced caching.

  • Built with Modular Architecture, RBAC, and JWT authentication using Access + Refresh token strategy.
  • Integrated Redis (Upstash) for OTP/session storage, sliding-window rate limiting, and 15x faster query caching.
TypeScriptNode.jsExpressMongoDBRedisBullMQSocket.ioCloudinarySSLCommerzJWTZod

ThoughtHub

Scalable Blog API Backend built with Modular Pattern for maintainability.

  • Developed robust RESTful API using Modular Pattern for improved maintainability.
  • Engineered advanced Search, Sort, and Filter using Mongoose query builders.
TypeScriptNode.jsExpressMongoDBMongooseJWTZod

Hotello

Hotel Booking Platform with real-time room availability and advanced filtering.

  • Built dynamic booking system with real-time room availability based on selected dates.
  • Implemented advanced filtering by price range and user ratings.
ReactNode.jsExpressMongoDBFirebaseTailwind CSS

Academic

Research

Undergraduate Thesis

Multimodal Skin Lesion Classification & Explainable AI

Dataset: HAM10000

Dataset

98.33%

Accuracy

0.9727

Macro F1

7

Classes

  • Developed a two-phase deep learning pipeline for classifying skin lesions into 7 categories using Swin Transformer.
  • Integrated clinical metadata (age, sex, localization) alongside image features to enhance classification reliability.
  • Implemented Supervised Contrastive Learning to learn robust and discriminative feature representations.
  • Applied Cross-Validation ensuring model robustness and generalization across the HAM10000 dataset.
  • Achieved 98.33% overall accuracy with a Macro F1-score of 0.9727, integrated with Explainable AI (XAI) for transparent diagnostic decision-making.
PythonPyTorchSwin TransformerCross-ValidationSupervised Contrastive LossCross-Entropy LossXAIHAM10000

Background

Education

B.Sc. in Computer Science and Engineering

Green University of Bangladesh

2022 – 2026CGPA: 3.74 / 4.00

Focusing on Software Engineering, Machine Learning, and Computer Vision. Conducting undergraduate thesis research on Multimodal Skin Lesion Classification using Swin Transformer and Explainable AI.

Higher Secondary Certificate (H.S.C.)

Govt. Tolaram College, Narayangonj, Dhaka

2020GPA: 5.00 / 5.00

Secondary School Certificate (S.S.C.)

I.E.T. Govt. High School,Narayangonj, Dhaka

2018GPA: 5.00 / 5.00

Say hello

Get in Touch

I'm currently open to new opportunities — whether it's a full-time position, freelance project, or just a conversation about tech. Drop me a message and I'll get back to you soon.

nahidulislam019045@gmail.com+880 1887 576888
Narayanganj, Dhaka, Bangladesh
GitHubLinkedIn