Transport Management Dashboard Built with Next.js, MongoDB, and Cloudinary

From UI to Full-Stack: 2 Weeks of Backend Learning! For the past two weeks, I’ve been learning the backend to understand how modern web applications function "under the hood." Today, I’m excited to share Scatch, a mini-project where I integrated my Frontend expertise with a secure, scalable Node.js backend. The Tech Stack & Challenges: # Backend: Built with Node.js & Express, using MongoDB for flexible data storage. # Authentication: Secured user accounts with Bcrypt hashing and handled session persistence with JWT and Cookies. # File Handling: Integrated Multer for dynamic product image uploads using memorystorage. # Frontend: Used EJS for server-side rendering, styled with Tailwind CSS for a responsive, modern look. # Security: Implemented Dotenv for environment variable management and custom middleware for route protection. Building this project helped me master the full CRUD lifecycle and understand the critical flow of data between the client and the server. It’s one thing to build a UI, it’s another to build the engine that powers it! 🔗 GitHub Repository: https://lnkd.in/g683J8mA If you find this project useful, please consider giving it a star! ⭐ I'd love to hear your feedback! #FullStackDeveloper #NodeJS #ExpressJS #MongoDB #WebDevelopment #MERNStack #TailwindCSS #PortfolioProject #LearningInPublic #BangladeshIT

The 3-second delay that was quietly killing user retention (and how we fixed it). Have you ever watched a beautifully designed app UI freeze while waiting for the backend to catch up? Recently, we ran into a bottleneck where a core dashboard was taking over 3 seconds to load. In the mobile world, 3 seconds might as well be a lifetime. Users were bouncing, and the experience felt sluggish. Instead of just throwing more server resources at the problem, I decided to tear down the request lifecycle and find the root cause. Here is how we turned a 3.2-second load time into a lightning-fast 200ms response: The Diagnosis: The issue wasn't the frontend UI. The Flutter app was rendering perfectly, but it was starving for data. Diving into the Laravel backend, the culprit became clear: the classic N+1 query problem, compounded by a lack of proper indexing on a rapidly growing PostgreSQL database. The Execution: Database Overhaul: I audited our PostgreSQL tables and implemented targeted composite indexes for the specific columns our heavy queries were filtering against. Backend Refactoring: I rewrote the underlying Laravel Eloquent relationships, replacing lazy loading with strict eager loading and raw SQL joins where the ORM was generating inefficient queries. Frontend Optimization: On the Flutter side, I implemented a more robust caching strategy and skeleton loaders so the user instantly felt progress the millisecond they opened the screen. The Result: API response times plummeted by over 90%. The dashboard now loads seamlessly, creating a significantly smoother user journey. Reduced database CPU load, saving us future scaling costs. As a full-stack developer, moments like this are why I love what I do. It’s not just about building features; it’s about writing code that respects the user's time and scales elegantly under the hood. #FullStackDevelopment #SoftwareEngineering #Laravel #Flutter #PostgreSQL #TechCaseStudy #WebDevelopment #Coding

🚀 I Haven’t Written a Single Line of Code Yet… But I Started Building a URL Shortener Instead of jumping straight into coding, I tried something different this time, I started with thinking. I asked myself: 👉 What actually happens when we paste a long URL and get a short one back? 👉 How does the system remember and redirect us instantly? That’s when I began sketching out a simple high-level design. At a basic level: A React frontend takes the long URL from the user A Node.js backend generates a unique short ID and handles redirection A MongoDB database stores the mapping between short and long URLs Then the flow clicked: User submits a long URL Backend generates a short ID Stores it in the database Returns a clean short URL When accessed → redirects to the original link Simple on the surface… but it opened up so many questions: How do we generate unique IDs at scale? How can we make redirects faster? (maybe caching 👀) What about tracking clicks or adding custom URLs? 📌 Biggest realization: Good systems aren’t just coded, they’re designed first. This is just the starting point. Next step: turning this design into a working backend ⚙️ Sharing my initial thought process below 👇 #SystemDesign #LearningInPublic #WebDevelopment #NodeJS #React #MongoDB

🚀 Understanding Backend API Architecture (MERN Stack) A clean and scalable backend is the backbone of any modern web application. Here’s a simple breakdown of how a well-structured API works: 🔹 Routes → Handle incoming requests 🔹 Controllers → Process requests & responses 🔹 Services → Contain business logic 🔹 Database → Store and manage data This layered architecture helps in: ✅ Better code organization ✅ Easy scalability ✅ Improved maintainability ✅ Clean separation of concerns As a MERN Stack Developer, I focus on building structured and scalable backend systems that can handle real-world applications efficiently. #MERN #BackendDevelopment #Nodejs #WebDevelopment #SoftwareEngineering #MongoDB #API #FullStackDeveloper

🚀 Built a Mini Chat App using Node.js, Express, MongoDB, Mongoose, and EJS — and this project turned out to be much more impactful than I expected. At first, it looked like a simple CRUD project. But while building it, I realized how much it helped me understand the real backend flow of a web app. What I implemented: Create new chats View all chats Edit existing messages Delete chats Dynamic EJS rendering Separate public folder for styling Clean route handling with Express MongoDB integration using Mongoose Method override for PUT & DELETE Timestamps for created and updated chats. What I actually learned: This project helped me understand: How routes connect with data How to design schemas/models with validations How CRUD operations work in a real app How form data flows through req.body Difference between render and redirect How backend, database, and templates all work together. The biggest takeaway for me: 👉 This wasn’t just about “making it work” 👉 It was about understanding why it works This project gave me a much stronger grasp of backend fundamentals, and I can already see how this will help me in my next bigger full-stack project: an Airbnb/WanderLust-style listing app. Still learning, still improving — but this one definitely felt like a solid step forward. 💻 #NodeJS #ExpressJS #MongoDB #Mongoose #EJS #WebDevelopment #BackendDevelopment #FullStackDevelopment #JavaScript #CRUD #LearningInPublic #100DaysOfCode

Searching for the exact same React and Express boilerplate setups again and again was getting incredibly tiring. Constantly context-switching to Google basic commands every time I started a project completely broke my flow, so I decided to finally do something about it. I built my own CLI Developer Assistant! 🤖💻 Instead of building another standard web app, I wanted to build a practical tool that actually solves a real annoyance and improves my daily workflow directly from the terminal. Here is how I built the architecture: 🧠 The Brain (Backend): A Node.js & Express API running locally, securely connected to a MongoDB Atlas database vault to store all my code snippets. ⌨️ The Walkie-Talkie (Client): A custom Command Line Interface (CLI) built with Node.js that uses fetch to talk to my API. My biggest "Aha!" moment: Stepping out of the browser and learning how to interact directly with the operating system was a massive mental shift. Learning how to use npm install -g to tap into the Windows OS PATH and register my own global bot command, along with parsing raw terminal inputs using process.argv, completely changed how I look at software engineering. Now, instead of breaking my flow to search the web, I just type: bot save "npm create vite@latest . -- --template react-ts" as "react-ts" ...and it instantly writes to my database. Later, bot get react-ts prints it right back out for me to use. I'm incredibly proud of how this decoupled system turned out! But I still need to copy and execute the commands manually—sometimes being a little lazy is the best motivation to learn something completely new! 😂 You can check out the source code here: [https://lnkd.in/gS49R2wj] #SoftwareEngineering #WebDevelopment #NodeJS #MongoDB #CLI #DeveloperTools #BuildInPublic

MERN vs. Next.js/Remix in 2026. Stop building "MERN" apps like it’s 2018. Most developers are still stuck in the "separate client, separate server" mindset. But the lines have blurred. If you aren't integrating these 3 things into your MERN stack in 2026, you're building legacy code: 1️⃣ Vector Databases: Using MongoDB? If you aren't using MongoDB Atlas Vector Search for AI-driven features, you're leaving UX on the table. 2️⃣ Serverless Functions: Express is great, but Node.js developers are moving toward hybrid architectures. Why manage a whole server for a simple auth check? 3️⃣ React Server Components (RSC): Yes, even in a pure MERN setup. Performance is no longer optional. The "M" in MERN now stands for Modern, not just Mongo. Is the traditional MERN stack dying, or just evolving? Let's discuss below. #MERNStack #WebDevelopment #ReactJS #NodeJS #SoftwareEngineering

𝗕𝗨𝗜𝗟𝗗𝗜𝗡𝗚 𝗔 𝗚𝗟𝗢𝗕𝗔𝗟 𝗣𝗘𝗧 𝗥𝗘𝗦𝗖𝗨𝗘 𝗣𝗟𝗔𝗧𝗙𝗢𝗥𝗠 As a developer, you want to solve real problems. One big issue is lost pets and stray animals needing a home. You can bridge this gap with technology. I built Patify, a platform to reunite lost pets with their families and facilitate adoptions. Here's what I used to build Patify: - Frontend: Next.js - Backend: Node.js - Database: MongoDB - Mobile: Flutter For a platform like this, Search Engine Optimization is key. Next.js allows us to use Server-Side Rendering. This means when a user creates a lost pet listing, the page is immediately indexable by search engine bots. Pet data is unstructured. MongoDB's document-based model allows us to store varying pet characteristics without rigid schemas. If you want to see how it works, you can check out Patify. Building a platform from scratch is challenging. But seeing the system connect a lost pet with its owner makes it worth it. I'd love to hear your thoughts on handling large-scale image rendering in Next.js or optimizing MongoDB geospatial queries. Source: https://lnkd.in/gYN9KpsD

🚀 Built & Deployed: URL Shortener – "URL SQUEEZER" Excited to share my latest full-stack project where I built a complete URL Shortener from scratch and deployed it to production. 🔗 Live Project: https://lnkd.in/gdYMSgsJ ⚠️ Note: First URL generation may take ~30 seconds (backend cold start on Render). 🔗 What it does: Converts long URLs into short, shareable links Redirects users seamlessly to the original URL Tracks URL mappings with database support 🛠️ Tech Stack: Frontend: React (Vite), Axios, CSS Backend: Node.js, Express.js Database: MongoDB Atlas Deployment: Vercel (Frontend) + Render (Backend) ⚙️ Key Features: Unique short ID generation Copy-to-clipboard functionality Input validation & error handling CORS-enabled API integration Environment-based configuration 🌍 What I learned: End-to-end MERN development Real-world deployment workflows Managing environment variables securely Connecting frontend ↔ backend in production Working with MongoDB Atlas (cloud DB) 💡 Next step: Adding analytics dashboard (click tracking, geo stats, etc.) Would love your feedback or suggestions! 🙌 #WebDevelopment #MERNStack #ReactJS #NodeJS #MongoDB #FullStack #Projects #Learning #Developers

🎥 Demo video attached showing the API working with the UI. 🚀 Book Store API with HTML UI I recently built a Book Store API with a simple UI to practice backend development and CRUD operations using Node.js,dotenv,Express, and MongoDB. 🔧 Tech Stack • Node.js • Express.js • MongoDB + Mongoose • HTML & CSS • Nodemon • Vercel (Deployment) ✨ Features • Add new books • View all books • Get book by ID • Update book details • Delete books • Simple HTML UI to interact with the API 📂 Project Links GitHub Repository: https://lnkd.in/g9ks_Vkf Live Demo on Vercel: https://lnkd.in/gEfD7ZAB 📚 What I Learned • Building REST APIs using Express • Using MongoDB with Mongoose • Connecting frontend UI with backend API • Deploying a project online #NodeJS #ExpressJS #MongoDB #BackendDevelopment #WebDevelopment #LearningInPublic