<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by SHOJOL ISLAM on Medium]]></title>
        <description><![CDATA[Stories by SHOJOL ISLAM on Medium]]></description>
        <link>https://medium.com/@devshojol?source=rss-688269b22109------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*N5nEP5AgzqT4_XeR</url>
            <title>Stories by SHOJOL ISLAM on Medium</title>
            <link>https://medium.com/@devshojol?source=rss-688269b22109------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Thu, 19 Nov 2026 19:48:18 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@devshojol/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[I Built a Free Interview Prep App for Developers in Bangladesh, in English and বাংলা]]></title>
            <link>https://medium.com/@devshojol/i-built-a-free-interview-prep-app-for-developers-in-bangladesh-in-english-and-%E0%A6%AC%E0%A6%BE%E0%A6%82%E0%A6%B2%E0%A6%BE-ad239f698ddf?source=rss-688269b22109------2</link>
            <guid isPermaLink="false">https://medium.com/p/ad239f698ddf</guid>
            <category><![CDATA[interview-preparation]]></category>
            <category><![CDATA[backend-interview]]></category>
            <category><![CDATA[web-development]]></category>
            <category><![CDATA[javascript]]></category>
            <category><![CDATA[react-interview-question]]></category>
            <dc:creator><![CDATA[SHOJOL ISLAM]]></dc:creator>
            <pubDate>Mon, 26 Oct 2026 09:22:58 GMT</pubDate>
            <atom:updated>2026-10-26T09:22:58.940Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*XN4rXvFK1aunbi6l6BxfKw.png" /></figure><p>A few weeks before an interview, most of us do the same thing. We open a folder of PDFs someone shared in a Facebook group (“Top 50 React Interview Questions”, “JS notes”), read the answers, nod, and feel ready.</p><p>Then the interviewer asks a follow-up.</p><p><em>“Okay, </em><em>useEffect runs after render. Why did your effect run twice?&quot;</em><br><em>&quot;You said </em><em>WHERE filters rows. So why can&#39;t you use </em><em>COUNT(*) in it?&quot;</em></p><p>A one-line answer you memorised doesn’t survive that second question. And if you think through technical ideas half in Bangla and half in English, like most developers here do, an English-only explanation is one more thing to translate in your head under pressure.</p><p>So I built <strong>Interview Prep BD</strong>: a free, bilingual practice app that helps you <em>say</em> the answer, not just read it.</p><p>👉 <strong>Try it:</strong> <a href="https://devshojol.github.io/Interview-prep-BD/">devshojol.github.io/Interview-prep-BD</a><br>👉 <strong>Source code:</strong> <a href="https://github.com/devshojol/Interview-prep-BD">github.com/devshojol/Interview-prep-BD</a></p><h3>What’s inside</h3><ul><li><strong>834 questions across 19 technologies:</strong> HTML, CSS, JavaScript, React, Next.js, <strong>React Native (with Expo and SQLite)</strong>, Node.js, Express, NestJS, MongoDB, Mongoose, PostgreSQL, SQL, Drizzle ORM, Redis, Vector Databases, Docker, NGINX and AWS.</li><li><strong>A behavioural &amp; projects section:</strong> “Tell me about yourself”, “Walk me through your project”, “What’s the hardest bug you’ve fixed?”, and more.</li><li><strong>Every question in English, বাংলা, or both side by side.</strong></li><li><strong>Four levels:</strong> Easy, Medium, Hard and Extreme, from viva-style fundamentals to senior system design.</li><li><strong>A voice interviewer</strong> that reads questions aloud in English, so practice sounds like the real thing.</li></ul><h3>1. A quick answer first, the deep answer below</h3><p>Every question has two layers.</p><p>The <strong>quick answer</strong> is 1–3 plain sentences: the thing you’d say first in an interview. It’s perfect for revising the night before.</p><p>The <strong>full answer</strong> below it explains <em>why</em> it works, with code, common mistakes, an interview tip and the follow-up questions an interviewer is likely to ask next. That’s the part that gets you through the second and third question.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*yF3FHEODcx2WbPgFMZZ5WA.png" /></figure><p><em>The quick answer on top, the full answer below, here in English and Bangla side by side.</em></p><h3>2. Real Bangla, not machine translation</h3><p>The Bangla is written the way developers here actually talk. Technical terms stay in English (closure, event loop, index), and the explanation around them is natural Bangla. You can switch between English, বাংলা and Both at any time, on desktop or phone.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*EHNQvG_ATnhrxL48im-CgA.png" /></figure><p><em>A quick-recall SQL question in Bangla on a phone.</em></p><h3>3. Quick recall for written tests and viva rounds</h3><p>Many interviews here start with an MCQ test or a “viva”, rapid fundamental questions. For that, each technology has <strong>10 quick-recall questions</strong> with short, plain answers and a tiny code example:</p><ul><li>What is hoisting, and how do var, let and const differ?</li><li>call vs apply vs bind?</li><li>WHERE vs HAVING?</li><li>app.use() vs app.get() in Express?</li><li>find() vs findOne() vs findById() in Mongoose?</li><li>What is Metro, and what is Hermes, in React Native?</li></ul><p>There’s a dedicated flashcard deck for them too. Shuffle and flip through 190 cards on the bus.</p><h3>4. Practise like it’s the real interview</h3><p>Reading answers feels productive, but it isn’t practice. The app is built around one habit: <strong>answer out loud first, then check.</strong></p><ul><li><strong>Interview mode</strong> hides the answer until you reveal it.</li><li>You <strong>grade yourself honestly</strong>: correct, partially correct or incorrect.</li><li><strong>MCQs are graded automatically.</strong></li><li><strong>Practice sessions</strong> (10, 20 or 30 questions) and a <strong>timed interview simulation</strong> that goes from easy to hard.</li><li>A <strong>daily challenge</strong> of three questions.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*9a39k6CFoEC82hPEHYDO0Q.png" /></figure><p><em>The dashboard shows your accuracy, weighted score, streak and what’s due for review.</em></p><h3>5. An interviewer that actually speaks</h3><p>Reading a question on a screen is not the same as hearing it from a person across the table. In the <strong>Interview Simulation</strong> and the <strong>Daily Challenge</strong>, an English voice now reads each question aloud (“Question 3…”), then hands over to you: <em>“Your turn: answer out loud.”</em></p><ul><li>Replay a question any time, or press <strong>V</strong>.</li><li><strong>Listen</strong> buttons on every question page, in practice sessions and on flashcards. Flashcards can read the question, then the quick answer when you flip.</li><li>Pick the voice and speed in Settings, and choose where questions are read automatically.</li><li>It uses your browser’s built-in voices, so there’s nothing to install. Chrome and Edge usually have the most natural-sounding ones.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ofRTmXHwpPaF1WwD0RC9hA.png" /></figure><p><em>The interviewer reads the question aloud, then it’s your turn to answer out loud.</em></p><h3>6. Your mistakes come back on a schedule</h3><p>Anything you get wrong or only half right comes back for review after 1 day, then 3, 7, 14 and 30 days. The app also spots <strong>weak topics</strong> (for example, React “Effects and Cleanup”) so you know what to study next. No more re-reading things you already know.</p><h3>7. Behavioural and project questions</h3><p>Technical rounds get the attention, but the first and last questions are often non-technical. The behavioural section tells you <strong>what the interviewer is checking</strong>, gives you a <strong>structure</strong> (like STAR), and a <strong>sample answer</strong> with blanks you fill in with your own story. No fake statistics, no copy-paste scripts.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*hzLCKOdt3vwKdA3DLs0RYg.png" /></figure><p><em>“Walk me through a project you built”: how to answer, a sample answer, common mistakes and likely follow-ups.</em></p><h3>8. React Native, including Expo and SQLite</h3><p>As a React Native developer myself, I wanted this section to cover what you actually do at work, not just “what is a component”. It includes:</p><ul><li>Core components, Flexbox, safe areas and the keyboard</li><li>FlatList and FlashList performance</li><li>Expo Router navigation and deep links</li><li>Choosing between AsyncStorage, SecureStore, MMKV and <strong>expo-sqlite</strong></li><li>SQLite migrations with SQLiteProvider, transactions, prepared statements and SQL injection</li><li>Offline-first sync</li><li>EAS Build and OTA updates (including what happens when an OTA update breaks native compatibility)</li><li>The New Architecture</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*mb6EwNMNQDZFoaLlitr-Sg.png" /></figure><p><em>“Where should each piece of data go?”: AsyncStorage vs SecureStore vs MMKV vs SQLite.</em></p><h3>Up to date, and checked</h3><p>Interview answers go stale fast. A 2022 answer about Next.js caching is simply wrong today. The content targets current versions (React 19.3, Next.js 16, Node.js 26 LTS, Express 5, React Native 0.88 with Expo SDK 58), labels legacy APIs as legacy, and links to the official docs for every answer.</p><p>Wherever it was possible, the code examples were actually <strong>run</strong>: JavaScript and Node.js snippets in Node, SQL queries in SQLite. The question bank also went through a full review pass for technical accuracy. There’s a <strong>Versions &amp; Sources</strong> page in the app that lists what each answer is based on.</p><p>If you spot a mistake, please open an issue on GitHub. That’s the whole point of making it open source.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*PrvxROJgUEoPnL5mqZIedA.png" /></figure><p><em>Pick a technology, or filter by area: frontend, mobile, backend, databases, DevOps &amp; cloud, AI &amp; search, career skills.</em></p><h3>A simple two-week plan</h3><p>If you have an interview coming up, here’s how I’d use it:</p><ol><li><strong>Days 1–3:</strong> Flip through the quick-recall flashcards for your stack. Mark anything you can’t explain.</li><li><strong>Days 4–10:</strong> One 20-question practice session a day in Interview mode. Answer out loud, then grade yourself honestly. Turn on the voice so you <em>hear</em> each question first.</li><li><strong>Every day:</strong> Clear your review queue first. Those are your weak spots.</li><li><strong>Days 11–13:</strong> Run the timed interview simulation, and practise the behavioural questions out loud.</li><li><strong>Day 14:</strong> Read only the quick answers for your weak topics. Sleep.</li></ol><h3>Free, private, no account</h3><p>It’s a plain HTML, CSS and JavaScript app with no backend and no login. Your progress stays in your browser, and you can export it as a backup. Want it offline? Download the repo and open index.html; it runs straight from your computer.</p><p>I’d love to hear what you think: which topics are missing, which answers helped, and what tripped you up in your last interview. Leave a comment, star the repo, or open an issue.</p><p><strong>Good luck with your next interview!</strong></p><p>👉 <a href="https://devshojol.github.io/Interview-prep-BD/">devshojol.github.io/Interview-prep-BD</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ad239f698ddf" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[What is the Model Context Protocol?]]></title>
            <link>https://medium.com/@devshojol/what-is-the-model-context-protocol-95bac8f10ed2?source=rss-688269b22109------2</link>
            <guid isPermaLink="false">https://medium.com/p/95bac8f10ed2</guid>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[mcp-server]]></category>
            <category><![CDATA[backend-engineering]]></category>
            <category><![CDATA[expressjs]]></category>
            <dc:creator><![CDATA[SHOJOL ISLAM]]></dc:creator>
            <pubDate>Fri, 07 Nov 2025 09:58:36 GMT</pubDate>
            <atom:updated>2025-11-07T09:58:36.317Z</atom:updated>
            <content:encoded><![CDATA[<h3>Understanding MCP Servers: A Beginner’s Guide</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/800/0*iIxXDZiaPAF0dxYg" /></figure><p>The Model Context Protocol (MCP) is an open standard that enables AI applications to securely connect with external data sources and tools. Think of it as a universal adapter that allows AI models like Claude to interact with your databases, APIs, file systems, and other services in a standardized way.</p><p>Before MCP, every integration between an AI model and an external service required custom code. MCP changes this by providing a common language that both AI applications and data sources can speak.</p><h3>Why MCP Matters</h3><p>Imagine you want your AI assistant to help you manage tasks in your project management tool, search through your company’s documentation, or analyze data from your database. Without MCP, developers would need to build custom integrations for each use case. With MCP, you write one server that exposes your data and tools, and any MCP-compatible AI application can use it.</p><p>The key benefits include:</p><p><strong>Standardization</strong>: One protocol works across different AI models and applications.</p><p><strong>Security</strong>: Built-in authentication and permission controls keep your data safe.</p><p><strong>Reusability</strong>: Write a server once, use it with multiple AI applications.</p><p><strong>Flexibility</strong>: Connect to databases, APIs, file systems, or any custom data source.</p><h3>How MCP Works</h3><p>The Model Context Protocol operates on a client-server architecture:</p><pre>┌─────────────────┐         ┌─────────────────┐         ┌─────────────────┐<br>│                 │         │                 │         │                 │<br>│   AI Model      │◄───────►│   MCP Server    │◄───────►│  Data Source    │<br>│   (Claude)      │   MCP   │                 │         │  (Database/API) │<br>│                 │         │                 │         │                 │<br>└─────────────────┘         └─────────────────┘         └─────────────────┘</pre><p>The AI model acts as the MCP client, sending requests to your MCP server. Your server handles these requests by interacting with your data sources and returning responses in a format the AI understands.</p><h3>Core Concepts</h3><h3>Resources</h3><p>Resources represent data that your server can provide. This could be file contents, database records, API responses, or any other information. Each resource has a unique URI and can be read by the AI model.</p><h3>Tools</h3><p>Tools are functions that the AI model can call to perform actions. Unlike resources which are read-only, tools can modify data, trigger workflows, or execute commands. For example, a tool might create a task, send an email, or update a database record.</p><h3>Prompts</h3><p>Prompts are reusable templates that help structure interactions. They provide context and instructions that guide how the AI should work with your data and tools.</p><h3>Building Your First MCP Server with Node.js and Express</h3><p>Let’s build a simple MCP server that manages a to-do list using Express.js. This server will expose both resources for reading tasks and tools for creating new ones, accessible via HTTP endpoints.</p><h3>Setting Up the Project</h3><p>First, create a new Node.js project and install the required dependencies:</p><pre>mkdir mcp-todo-server<br>cd mcp-todo-server<br>npm init -y<br>npm install @modelcontextprotocol/sdk express cors</pre><p>We’re installing Express for the HTTP server and CORS to handle cross-origin requests safely.</p><h3>Creating the Server</h3><p>Create a file named server.js:</p><pre>import { Server } from &quot;@modelcontextprotocol/sdk/server/index.js&quot;;<br>import { SSEServerTransport } from &quot;@modelcontextprotocol/sdk/server/sse.js&quot;;<br>import {<br>  CallToolRequestSchema,<br>  ListResourcesRequestSchema,<br>  ListToolsRequestSchema,<br>  ReadResourceRequestSchema,<br>} from &quot;@modelcontextprotocol/sdk/types.js&quot;;<br>import express from &quot;express&quot;;<br>import cors from &quot;cors&quot;;<br><br>// In-memory task storage<br>const tasks = [<br>  { id: 1, title: &quot;Learn MCP&quot;, completed: false },<br>  { id: 2, title: &quot;Build a server&quot;, completed: false }<br>];<br>let nextId = 3;<br>// Create Express app<br>const app = express();<br>app.use(cors());<br>app.use(express.json());<br>// Create MCP server instance<br>const server = new Server(<br>  {<br>    name: &quot;todo-server&quot;,<br>    version: &quot;1.0.0&quot;,<br>  },<br>  {<br>    capabilities: {<br>      resources: {},<br>      tools: {},<br>    },<br>  }<br>);<br>// Handle resource listing<br>server.setRequestHandler(ListResourcesRequestSchema, async () =&gt; {<br>  return {<br>    resources: [<br>      {<br>        uri: &quot;todo://tasks&quot;,<br>        name: &quot;All Tasks&quot;,<br>        mimeType: &quot;application/json&quot;,<br>        description: &quot;List of all tasks&quot;,<br>      },<br>    ],<br>  };<br>});<br>// Handle resource reading<br>server.setRequestHandler(ReadResourceRequestSchema, async (request) =&gt; {<br>  if (request.params.uri === &quot;todo://tasks&quot;) {<br>    return {<br>      contents: [<br>        {<br>          uri: &quot;todo://tasks&quot;,<br>          mimeType: &quot;application/json&quot;,<br>          text: JSON.stringify(tasks, null, 2),<br>        },<br>      ],<br>    };<br>  }<br>  throw new Error(&quot;Resource not found&quot;);<br>});<br>// Handle tool listing<br>server.setRequestHandler(ListToolsRequestSchema, async () =&gt; {<br>  return {<br>    tools: [<br>      {<br>        name: &quot;create_task&quot;,<br>        description: &quot;Create a new task&quot;,<br>        inputSchema: {<br>          type: &quot;object&quot;,<br>          properties: {<br>            title: {<br>              type: &quot;string&quot;,<br>              description: &quot;Title of the task&quot;,<br>            },<br>          },<br>          required: [&quot;title&quot;],<br>        },<br>      },<br>      {<br>        name: &quot;complete_task&quot;,<br>        description: &quot;Mark a task as completed&quot;,<br>        inputSchema: {<br>          type: &quot;object&quot;,<br>          properties: {<br>            id: {<br>              type: &quot;number&quot;,<br>              description: &quot;ID of the task to complete&quot;,<br>            },<br>          },<br>          required: [&quot;id&quot;],<br>        },<br>      },<br>    ],<br>  };<br>});<br>// Handle tool execution<br>server.setRequestHandler(CallToolRequestSchema, async (request) =&gt; {<br>  if (request.params.name === &quot;create_task&quot;) {<br>    const title = request.params.arguments.title;<br>    const newTask = { id: nextId++, title, completed: false };<br>    tasks.push(newTask);<br>    <br>    return {<br>      content: [<br>        {<br>          type: &quot;text&quot;,<br>          text: `Created task: ${title} (ID: ${newTask.id})`,<br>        },<br>      ],<br>    };<br>  }<br>  <br>  if (request.params.name === &quot;complete_task&quot;) {<br>    const id = request.params.arguments.id;<br>    const task = tasks.find(t =&gt; t.id === id);<br>    <br>    if (!task) {<br>      throw new Error(`Task ${id} not found`);<br>    }<br>    <br>    task.completed = true;<br>    <br>    return {<br>      content: [<br>        {<br>          type: &quot;text&quot;,<br>          text: `Marked task ${id} as completed`,<br>        },<br>      ],<br>    };<br>  }<br>  <br>  throw new Error(&quot;Unknown tool&quot;);<br>});<br>// Health check endpoint<br>app.get(&quot;/health&quot;, (req, res) =&gt; {<br>  res.json({ status: &quot;healthy&quot;, timestamp: new Date().toISOString() });<br>});<br>// MCP endpoint using Server-Sent Events<br>app.get(&quot;/sse&quot;, async (req, res) =&gt; {<br>  console.log(&quot;New SSE connection established&quot;);<br>  <br>  const transport = new SSEServerTransport(&quot;/message&quot;, res);<br>  await server.connect(transport);<br>  <br>  req.on(&quot;close&quot;, () =&gt; {<br>    console.log(&quot;SSE connection closed&quot;);<br>  });<br>});<br>// Message handling endpoint<br>app.post(&quot;/message&quot;, async (req, res) =&gt; {<br>  console.log(&quot;Received message:&quot;, req.body);<br>  // The SSE transport handles the message routing<br>  res.sendStatus(200);<br>});<br>// Optional: REST API endpoints for direct access (non-MCP)<br>app.get(&quot;/api/tasks&quot;, (req, res) =&gt; {<br>  res.json(tasks);<br>});<br>app.post(&quot;/api/tasks&quot;, (req, res) =&gt; {<br>  const { title } = req.body;<br>  if (!title) {<br>    return res.status(400).json({ error: &quot;Title is required&quot; });<br>  }<br>  <br>  const newTask = { id: nextId++, title, completed: false };<br>  tasks.push(newTask);<br>  res.status(201).json(newTask);<br>});<br>app.patch(&quot;/api/tasks/:id/complete&quot;, (req, res) =&gt; {<br>  const id = parseInt(req.params.id);<br>  const task = tasks.find(t =&gt; t.id === id);<br>  <br>  if (!task) {<br>    return res.status(404).json({ error: &quot;Task not found&quot; });<br>  }<br>  <br>  task.completed = true;<br>  res.json(task);<br>});<br>// Start the server<br>const PORT = process.env.PORT || 3000;<br>app.listen(PORT, () =&gt; {<br>  console.log(`MCP Todo Server running on http://localhost:${PORT}`);<br>  console.log(`Health check: http://localhost:${PORT}/health`);<br>  console.log(`MCP endpoint: http://localhost:${PORT}/sse`);<br>  console.log(`REST API: http://localhost:${PORT}/api/tasks`);<br>});</pre><h3>Understanding the Code</h3><p>The server combines Express.js HTTP functionality with MCP protocol handling:</p><p><strong>Express Setup</strong>: We create an Express app with CORS enabled and JSON parsing middleware for handling HTTP requests.</p><p><strong>MCP Handlers</strong>: The same four request handlers from before, but now accessible via HTTP instead of stdio.</p><p><strong>SSE Transport</strong>: Server-Sent Events (SSE) provide real-time communication between the MCP client and server. The /sse endpoint establishes the connection.</p><p><strong>Hybrid Approach</strong>: We include both MCP endpoints and traditional REST API endpoints, giving you flexibility in how you interact with the server.</p><p><strong>Health Check</strong>: The /health endpoint allows you to verify the server is running and responsive.</p><h3>Running the Server</h3><p>Update your package.json to use ES modules:</p><pre>{<br>  &quot;type&quot;: &quot;module&quot;,<br>  &quot;scripts&quot;: {<br>    &quot;start&quot;: &quot;node server.js&quot;,<br>    &quot;dev&quot;: &quot;nodemon server.js&quot;<br>  }<br>}</pre><p>Run the server:</p><pre>npm start</pre><p>You should see output like:</p><pre>MCP Todo Server running on http://localhost:3000<br>Health check: http://localhost:3000/health<br>MCP endpoint: http://localhost:3000/sse<br>REST API: http://localhost:3000/api/tasks</pre><h3>Testing the Server</h3><p>You can test the traditional REST endpoints using curl or any HTTP client:</p><pre># Get all tasks<br>curl http://localhost:3000/api/tasks</pre><pre># Create a new task<br>curl -X POST <a href="http://localhost:3000/api/tasks">http://localhost:3000/api/tasks</a> \<br>  -H &quot;Content-Type: application/json&quot; \<br>  -d &#39;{&quot;title&quot;: &quot;Test the MCP server&quot;}&#39;</pre><pre># Complete a task<br>curl -X PATCH <a href="http://localhost:3000/api/tasks/1/complete">http://localhost:3000/api/tasks/1/complete</a></pre><pre># Health check<br>curl <a href="http://localhost:3000/health">http://localhost:3000/health</a></pre><h3>Connecting to Claude</h3><p>To use your Express-based MCP server with Claude Desktop, you need to configure it in the Claude settings. Add this to your Claude Desktop configuration file:</p><p>For macOS: ~/Library/Application Support/Claude/claude_desktop_config.json</p><p>For Windows: %APPDATA%\Claude\claude_desktop_config.json</p><pre>{<br>  &quot;mcpServers&quot;: {<br>    &quot;todo&quot;: {<br>      &quot;url&quot;: &quot;http://localhost:3000/sse&quot;<br>    }<br>  }<br>}</pre><p>After restarting Claude Desktop, your server will be available. You can ask Claude to read your tasks or create new ones, and it will use your MCP server behind the scenes.</p><h3>Why Express.js?</h3><p>Using Express.js for your MCP server provides several advantages:</p><p><strong>Familiar Framework</strong>: If you’ve built web applications before, Express.js is a well-known and comfortable environment.</p><p><strong>HTTP-Based</strong>: Unlike stdio transport, HTTP allows your server to be accessed over the network, making it more flexible for deployment.</p><p><strong>Multiple Clients</strong>: An Express server can handle multiple simultaneous connections, letting different applications use your MCP server at the same time.</p><p><strong>REST API Bonus</strong>: You can expose both MCP protocol endpoints and traditional REST APIs, giving you backward compatibility with existing tools.</p><p><strong>Middleware Ecosystem</strong>: Express’s rich middleware ecosystem lets you easily add authentication, logging, rate limiting, and other features.</p><p><strong>Production Ready</strong>: Express is battle-tested and ready for production deployments with proper scaling and monitoring.</p><h3>Architecture Deep Dive</h3><p>Here’s how the complete flow works when you ask Claude to create a task:</p><pre>User: &quot;Create a task to review code&quot;<br>           │<br>           ▼<br>    ┌──────────────┐<br>    │    Claude    │<br>    │   (Client)   │<br>    └──────┬───────┘<br>           │ 1. HTTP Request to /sse endpoint<br>           │<br>           ▼<br>    ┌──────────────┐<br>    │   Express    │<br>    │  MCP Server  │<br>    └──────┬───────┘<br>           │ 2. MCP processes tool call<br>           │ 3. Adds task to array<br>           │<br>           ▼<br>    ┌──────────────┐<br>    │   Returns    │<br>    │   Response   │<br>    │   via SSE    │<br>    └──────────────┘<br>           │<br>           ▼<br>    Claude confirms: &quot;I&#39;ve created a task to review code&quot;</pre><h3>Best Practices for Express-Based MCP Servers</h3><p><strong>Error Handling</strong>: Always validate inputs and provide clear error messages. Add Express error-handling middleware to catch and format errors consistently.</p><pre>app.use((err, req, res, next) =&gt; {<br>  console.error(err);<br>  res.status(500).json({ error: err.message });<br>});</pre><p><strong>Resource URIs</strong>: Use descriptive, hierarchical URIs like todo://tasks/1 or database://users. This makes it easier to understand what each resource represents.</p><p><strong>Tool Design</strong>: Keep tools focused on single actions. Instead of one complex tool, create multiple simple tools that can be combined.</p><p><strong>Authentication</strong>: For production servers, implement proper authentication. Express makes it easy to add JWT tokens, API keys, or OAuth:</p><pre>import jwt from &#39;jsonwebtoken&#39;;</pre><pre>app.use(&#39;/sse&#39;, (req, res, next) =&gt; {<br>  const token = req.headers.authorization?.split(&#39; &#39;)[1];<br>  if (!token) return res.status(401).json({ error: &#39;Unauthorized&#39; });<br>  <br>  try {<br>    jwt.verify(token, process.env.JWT_SECRET);<br>    next();<br>  } catch {<br>    res.status(403).json({ error: &#39;Invalid token&#39; });<br>  }<br>});</pre><p><strong>Logging</strong>: Add comprehensive logging to debug issues. Use libraries like Winston or Morgan for production-grade logging.</p><p><strong>Rate Limiting</strong>: Protect your server from abuse by adding rate limiting:</p><pre>import rateLimit from &#39;express-rate-limit&#39;;<br><br>const limiter = rateLimit({<br>  windowMs: 15 * 60 * 1000, // 15 minutes<br>  max: 100 // limit each IP to 100 requests per windowMs<br>});<br>app.use(&#39;/api/&#39;, limiter);</pre><p><strong>Environment Configuration</strong>: Use environment variables for configuration:</p><pre>const PORT = process.env.PORT || 3000;<br>const NODE_ENV = process.env.NODE_ENV || &#39;development&#39;;</pre><h3>Real-World Use Cases</h3><p><strong>Database Access</strong>: Create a server that lets Claude query your database using natural language. The AI can translate requests into SQL and return formatted results.</p><p><strong>File System Navigation</strong>: Build a server that provides safe, sandboxed access to specific directories, allowing Claude to read, search, and analyze files.</p><p><strong>API Integration</strong>: Wrap third-party APIs in an MCP server, giving Claude the ability to interact with services like Slack, Jira, or GitHub.</p><p><strong>Custom Tools</strong>: Expose company-specific tools and workflows, letting Claude help with internal processes while maintaining security boundaries.</p><h3>Next Steps</h3><p>Now that you understand the basics of MCP servers, here are some ways to expand your knowledge:</p><p>Explore the official MCP documentation at modelcontextprotocol.io for advanced features like streaming responses and complex authentication.</p><p>Study existing MCP servers in the community. Many open-source implementations are available on GitHub.</p><p>Build a server for your own use case. The best way to learn is by solving real problems.</p><p>Experiment with different transport methods. While we used stdio, MCP also supports HTTP and other protocols.</p><h3>Conclusion</h3><p>The Model Context Protocol represents a significant step forward in making AI assistants more useful and integrated with our existing tools and data. By standardizing how AI models connect to external services, MCP makes it easier for developers to build powerful, context-aware AI applications.</p><p>Starting with a simple Node.js server like our to-do example, you can gradually build more sophisticated integrations that bring real value to your workflows. The key is understanding the three core concepts — resources, tools, and prompts — and how they work together to create a seamless experience.</p><p>As the MCP ecosystem grows, we’ll see more pre-built servers, better tooling, and wider adoption across AI applications. Getting started now means you’ll be ready to take advantage of this powerful standard as it evolves.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=95bac8f10ed2" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Building Better Loading States: Introducing React Native Smart Placeholder]]></title>
            <link>https://medium.com/@devshojol/building-better-loading-states-introducing-react-native-smart-placeholder-3ccd8d6f30ef?source=rss-688269b22109------2</link>
            <guid isPermaLink="false">https://medium.com/p/3ccd8d6f30ef</guid>
            <category><![CDATA[npm-package]]></category>
            <category><![CDATA[expo]]></category>
            <category><![CDATA[react-native]]></category>
            <category><![CDATA[react-native-developers]]></category>
            <dc:creator><![CDATA[SHOJOL ISLAM]]></dc:creator>
            <pubDate>Fri, 17 Oct 2025 11:30:00 GMT</pubDate>
            <atom:updated>2025-11-07T11:00:59.144Z</atom:updated>
            <content:encoded><![CDATA[<p>As mobile developers, we’ve all been there — staring at a blank screen while data loads, wondering how to keep users engaged during those crucial first moments. The solution? Smart, beautiful placeholder animations that set the right expectations and create a polished user experience.</p><p>Today, I’m excited to share <strong>React Native Smart Placeholder</strong> — a lightweight, zero-dependency library that brings professional loading states to your React Native apps with minimal effort.</p><h3>Why Another Placeholder Library?</h3><p>Fair question. While there are skeleton loaders out there, I found myself repeatedly facing the same challenges:</p><ul><li><strong>Heavy dependencies</strong> that bloat the app bundle</li><li><strong>Limited animation options</strong> that don’t match modern design trends</li><li><strong>Complex APIs</strong> that require pages of documentation just to show a simple loading state</li><li><strong>Poor TypeScript support</strong> leading to runtime surprises</li></ul><p>React Native Smart Placeholder solves these pain points with a simple philosophy: maximum flexibility, minimal complexity.</p><h3>What Makes It Smart?</h3><h3>Seven Animation Styles Out of the Box</h3><p>Different UI elements deserve different loading treatments. That’s why the library includes seven carefully crafted animation styles:</p><ol><li><strong>Linear</strong> — The classic left-to-right shimmer effect</li><li><strong>Radial</strong> — A circular wave that expands from center</li><li><strong>Directional</strong> — Vertical top-to-bottom animation</li><li><strong>Reverse</strong> — Right-to-left for RTL-friendly interfaces</li><li><strong>Pulse</strong> — Gentle breathing effect for subtle elegance</li><li><strong>Fade</strong> — Simple opacity transitions</li><li><strong>Skeleton</strong> — Static placeholder when animation isn’t needed</li></ol><h3>Zero External Dependencies</h3><p>Built entirely on React Native’s native Animated API, it adds virtually no weight to your app. No heavy animation libraries, no unnecessary bloat — just pure, optimized animations.</p><h3>Dead Simple API</h3><p>Here’s all you need to create a beautiful loading card:</p><pre>&lt;SmartPlaceholder <br>  width=&quot;100%&quot; <br>  height={80} <br>  borderRadius={10} <br>  animationStyle=&quot;linear&quot; <br>/&gt;</pre><p>That’s it. No complex configuration, no boilerplate, no headaches.</p><h3>Real-World Examples</h3><h3>Loading a User Profile</h3><pre>&lt;View style={styles.profileContainer}&gt;<br>  &lt;SmartPlaceholder <br>    width={80} <br>    height={80} <br>    borderRadius={40} <br>    animationStyle=&quot;pulse&quot; <br>  /&gt;<br>  &lt;View style={styles.profileInfo}&gt;<br>    &lt;SmartPlaceholder <br>      width=&quot;80%&quot; <br>      height={20} <br>      animationStyle=&quot;linear&quot; <br>    /&gt;<br>    &lt;SmartPlaceholder <br>      width=&quot;60%&quot; <br>      height={16} <br>      animationStyle=&quot;linear&quot; <br>    /&gt;<br>  &lt;/View&gt;<br>&lt;/View&gt;</pre><h3>Loading a List of Cards</h3><pre>{[1, 2, 3, 4].map((item) =&gt; (<br>  &lt;SmartPlaceholder<br>    key={item}<br>    width=&quot;100%&quot;<br>    height={120}<br>    borderRadius={12}<br>    animationStyle=&quot;radial&quot;<br>    backgroundColor=&quot;#E8E8E8&quot;<br>    animationColor=&quot;#F5F5F5&quot;<br>  /&gt;<br>))}</pre><h3>Dark Mode Support</h3><pre>&lt;SmartPlaceholder<br>  width=&quot;100%&quot;<br>  height={60}<br>  borderRadius={8}<br>  backgroundColor=&quot;#2A2A2A&quot;<br>  animationColor=&quot;#3F3F3F&quot;<br>  animationStyle=&quot;fade&quot;<br>/&gt;</pre><h3>Performance Considerations</h3><p>Loading states might seem simple, but they can impact performance if not done right. React Native Smart Placeholder is built with performance in mind:</p><ul><li><strong>Native Animated API</strong> ensures 60 FPS animations that run on the UI thread</li><li><strong>Optimized interpolations</strong> minimize JavaScript bridge traffic</li><li><strong>Memoized components</strong> prevent unnecessary re-renders</li><li><strong>Lightweight bundle</strong> keeps your app size in check</li></ul><h3>Customization Power</h3><p>While the defaults work great, you have complete control when you need it:</p><pre>&lt;SmartPlaceholder<br>  width={200}<br>  height={40}<br>  borderRadius={20}<br>  backgroundColor=&quot;#CUSTOM_BASE&quot;<br>  animationColor=&quot;#CUSTOM_HIGHLIGHT&quot;<br>  animationStyle=&quot;directional&quot;<br>  style={{ marginVertical: 10 }}<br>/&gt;</pre><p>Every prop is optional. Every style is customizable. The library adapts to your design system, not the other way around.</p><h3>TypeScript First</h3><p>Full TypeScript support means you get autocomplete and type safety out of the box:</p><pre>interface PlaceholderProps {<br>  width?: number | string;<br>  height?: number | string;<br>  backgroundColor?: string;<br>  animationColor?: string;<br>  animationStyle?: &#39;linear&#39; | &#39;radial&#39; | &#39;directional&#39; | <br>                   &#39;reverse&#39; | &#39;pulse&#39; | &#39;fade&#39; | &#39;skeleton&#39;;<br>  borderRadius?: number;<br>  style?: StyleProp&lt;ViewStyle&gt;;<br>}</pre><h3>Getting Started</h3><p>Installation is as simple as:</p><pre>npm install react-native-smart-placeholder</pre><p>Or if you prefer Yarn:</p><pre>yarn add react-native-smart-placeholder</pre><p>Works with both Expo and React Native CLI projects. No linking, no native code, no hassle.</p><h3>The Philosophy Behind the Library</h3><p>Creating React Native Smart Placeholder taught me an important lesson about open source: <strong>solve one problem exceptionally well</strong>.</p><p>This library doesn’t try to be everything. It doesn’t bundle form validators, state management, or a kitchen sink. It does one thing — beautiful loading placeholders — and does it really well.</p><p>That focus means:</p><ul><li>Smaller bundle size for your app</li><li>Easier to learn and use</li><li>More reliable and maintainable</li><li>Better performance</li></ul><h3>What’s Next?</h3><p>The library is production-ready today, but the roadmap includes:</p><ul><li><strong>More animation presets</strong> based on popular design systems</li><li><strong>Composite placeholders</strong> for common patterns (profile cards, list items, etc.)</li><li><strong>Animation customization</strong> for duration and easing</li><li><strong>Accessibility improvements</strong> with better screen reader support</li></ul><h3>Try It Today</h3><p>Check out the package on npm: <a href="https://www.npmjs.com/package/react-native-smart-placeholder">react-native-smart-placeholder</a></p><p>Browse the code on GitHub: <a href="https://github.com/devshojol/react-native-smart-placeholder">react-native-smart-placeholder</a></p><p>I’d love to hear your feedback, see how you’re using it, and welcome contributions from the community.</p><h3>Final Thoughts</h3><p>Loading states are often an afterthought in app development, but they’re one of the first things users experience. A polished loading animation shows attention to detail and sets expectations for the quality of your entire app.</p><p>React Native Smart Placeholder makes it trivial to add that polish. No excuses, no complexity — just beautiful loading states in seconds.</p><p>Give it a try in your next project, and let me know what you think!</p><p><em>Have questions or suggestions? Drop them in the comments below or open an issue on GitHub. Happy coding!</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=3ccd8d6f30ef" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Understanding 2D Arrays in C: A Beginner-Friendly Guide 2025.]]></title>
            <link>https://medium.com/@devshojol/understanding-2d-arrays-in-c-a-beginner-friendly-guide-2025-7c156aa19225?source=rss-688269b22109------2</link>
            <guid isPermaLink="false">https://medium.com/p/7c156aa19225</guid>
            <category><![CDATA[matrix]]></category>
            <category><![CDATA[beginners-guide]]></category>
            <category><![CDATA[cprogrammingapp]]></category>
            <category><![CDATA[2d-arrays]]></category>
            <dc:creator><![CDATA[SHOJOL ISLAM]]></dc:creator>
            <pubDate>Tue, 13 May 2025 09:11:50 GMT</pubDate>
            <atom:updated>2025-05-13T09:11:50.016Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/361/1*-hIN4enanTHWjsQXMZGFFQ.jpeg" /></figure><h3>📌 Introduction: Why Do We Need 2D Arrays in C?</h3><p>When you’re working with C, you quickly realize that it lacks higher-level data structures like <strong>objects</strong> found in languages like JavaScript or Python.</p><p>In JavaScript, if you wanted to store similar data, you might use an array of objects:</p><pre>const students = [<br>  { name: &quot;Alice&quot;, age: 20 },<br>  { name: &quot;Bob&quot;, age: 22 }<br>];</pre><p>However, in C, <strong>there are no objects</strong> by default.</p><p>If you want to group related data (like student marks in different subjects), the closest basic structure you have is an <strong>array</strong>, and if the data needs two dimensions (rows and columns), you use a <strong>2D array</strong>.</p><p>Example:</p><pre>int marks[2][3] = {<br>  {85, 90, 92},  // Student 1&#39;s marks<br>  {78, 88, 84}   // Student 2&#39;s marks<br>};</pre><p>Here:</p><ul><li><strong>Rows</strong> represent different students.</li><li><strong>Columns</strong> represent marks in different subjects.</li></ul><p><strong>In short</strong>:<br> 👉 2D arrays help simulate table-like data structures where simple arrays or variables are not enough.</p><h3>📌 How Input and Output Work in 2D Arrays</h3><p>A <strong>2D array</strong> in C is just an <strong>array of arrays</strong>.<br>You usually take input using <strong>nested loops</strong>.</p><p><strong>Example: Input and Output for a 2x3 array</strong></p><pre>#include &lt;stdio.h&gt;<br><br>int main() {<br>    int arr[2][3];<br>    <br>    // Input<br>    for(int i = 0; i &lt; 2; i++) {<br>        for(int j = 0; j &lt; 3; j++) {<br>            printf(&quot;Enter value for arr[%d][%d]: &quot;, i, j);<br>            scanf(&quot;%d&quot;, &amp;arr[i][j]);<br>        }<br>    }<br><br>    // Output<br>    printf(&quot;\nMatrix:\n&quot;);<br>    for(int i = 0; i &lt; 2; i++) {<br>        for(int j = 0; j &lt; 3; j++) {<br>            printf(&quot;%d &quot;, arr[i][j]);<br>        }<br>        printf(&quot;\n&quot;);<br>    }<br>    <br>    return 0;<br>}</pre><p>✅ Notice how:</p><ul><li>Outer loop → rows</li><li>Inner loop → columns</li></ul><h3>📌 How to Print a Single Row or Column</h3><h3>Printing a Single Row</h3><p>Example: Print row 1 (index 0):</p><pre>for(int j = 0; j &lt; 3; j++) {<br>    printf(&quot;%d &quot;, arr[0][j]);<br>}</pre><p>✅ Here, the <strong>row is fixed</strong>, and the <strong>columns vary</strong>.</p><h3>Printing a Single Column</h3><p>Example: Print column 2 (index 1):</p><pre>for(int i = 0; i &lt; 2; i++) {<br>    printf(&quot;%d\n&quot;, arr[i][1]);<br>}</pre><p>✅ Here, the <strong>column is fixed</strong>, and the <strong>rows vary</strong>.</p><h3>📌 Types of Matrices (with Simple Definitions and Examples)</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*IfgGDmno8s18qRqnGduTCw.png" /></figure><h3>📌 How to Check Matrix Types in C</h3><p>Here are <strong>basic C snippets</strong> to check each matrix type:</p><h3>1. Check Zero Matrix</h3><pre>int isZero = 1;<br>for(int i = 0; i &lt; rows; i++) {<br>    for(int j = 0; j &lt; cols; j++) {<br>        if(arr[i][j] != 0) {<br>            isZero = 0;<br>            break;<br>        }<br>    }<br>}</pre><h3>2. Check Diagonal Matrix</h3><pre>int isDiagonal = 1;<br>for(int i = 0; i &lt; rows; i++) {<br>    for(int j = 0; j &lt; cols; j++) {<br>        if(i != j &amp;&amp; arr[i][j] != 0) {<br>            isDiagonal = 0;<br>        }<br>    }<br>}</pre><h3>3. Check Scalar Matrix</h3><pre>int isScalar = 1;<br>int diagVal = arr[0][0];<br>for(int i = 0; i &lt; rows; i++) {<br>    for(int j = 0; j &lt; cols; j++) {<br>        if((i == j &amp;&amp; arr[i][j] != diagVal) || (i != j &amp;&amp; arr[i][j] != 0)) {<br>            isScalar = 0;<br>        }<br>    }<br>}</pre><h3>4. Check Unit (Identity) Matrix</h3><pre>int isUnit = 1;<br>for(int i = 0; i &lt; rows; i++) {<br>    for(int j = 0; j &lt; cols; j++) {<br>        if((i == j &amp;&amp; arr[i][j] != 1) || (i != j &amp;&amp; arr[i][j] != 0)) {<br>            isUnit = 0;<br>        }<br>    }<br>}</pre><h3>📌 Practice Problems for Beginners</h3><p>Here are a few beginner-friendly questions you can try:</p><ol><li><strong>Take a 3x3 matrix and print its diagonal elements.</strong></li><li><strong>Check whether a given matrix is a Zero Matrix or not.</strong></li><li><strong>Transpose a 2D matrix (convert rows into columns).</strong></li><li><strong>Sum of all elements in a particular row or column.</strong></li><li><strong>Check if a matrix is a Row Matrix or a Column Matrix.</strong></li><li><strong>Print the upper and lower triangles of a square matrix.</strong></li><li><strong>Find the largest element in each row of a matrix.</strong></li></ol><h3>🎯 Conclusion</h3><p>2D arrays are one of the most important tools in C programming for handling structured data like grids, tables, or matrices.</p><p>Understanding <strong>input/output</strong>, <strong>row/column access</strong>, <strong>different matrix types</strong>, and <strong>checking matrix properties</strong> will build a strong foundation — not just for coding but also for <strong>competitive programming (CP)</strong> and <strong>data structures</strong>.</p><p>Happy coding! 🚀</p><p>#C #Programming #CProgramming #CompetitiveProgramming #Matrix #2DArrays #BeginnerFriendly #CodeNewbie #Learning</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=7c156aa19225" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[ Build Desktop Apps Faster with Electron.js]]></title>
            <link>https://medium.com/@devshojol/build-desktop-apps-faster-with-electron-js-react-tailwindcss-react-router-boilerplate-b84ffb9de1d8?source=rss-688269b22109------2</link>
            <guid isPermaLink="false">https://medium.com/p/b84ffb9de1d8</guid>
            <category><![CDATA[tailwind-css]]></category>
            <category><![CDATA[react-dom]]></category>
            <category><![CDATA[electronjs]]></category>
            <category><![CDATA[reactjs]]></category>
            <dc:creator><![CDATA[SHOJOL ISLAM]]></dc:creator>
            <pubDate>Mon, 12 May 2025 11:35:23 GMT</pubDate>
            <atom:updated>2025-05-12T13:15:48.791Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/985/1*aZnntHcPj5zWZ-pEXj8ZyQ.png" /></figure><h3>🚀 Build Desktop Apps Faster with Electron.js + React + TailwindCSS + React Router Boilerplate 2025 SHOJOL ISLAM.</h3><h3>Introduction</h3><p>Setting up a new Electron app is often time-consuming. Configuring Vite, integrating React, setting up TailwindCSS, and getting routing right — it can easily take hours before you even start coding your real app.</p><p>I felt this pain many times while building desktop apps, so I decided to create a modern, minimal, and super-fast <strong>Electron.js + React + TailwindCSS + React Router Boilerplate</strong>! 🎯</p><p>This starter kit lets you <strong>focus on building amazing apps</strong>, not on wiring up basic things.</p><h3>📦 What’s Inside?</h3><p>Here’s what you get out of the box:</p><ul><li><strong>Electron.js</strong> — Desktop application framework.</li><li><strong>React.js</strong> — Frontend library.</li><li><strong>Vite</strong> — Lightning-fast development and build tool.</li><li><strong>TailwindCSS v4</strong> — Utility-first CSS framework.</li><li><strong>React Router DOM</strong> — SPA routing for React.</li><li><strong>ESLint + Prettier</strong> — Clean and consistent code style.</li></ul><p>It’s lightweight, scalable, and easy to extend.</p><h3>✨ Features</h3><p>✅ Blazing-fast React + Vite development<br> ✅ Preconfigured TailwindCSS v4 for styling<br> ✅ React Router DOM for easy page routing<br> ✅ Electron main process is ready to go<br> ✅ Hot Module Replacement (HMR) — frontend reloads instantly<br> ✅ Production build configuration included<br> ✅ Clean code enforced with ESLint and Prettier</p><h3>🚀 Getting Started</h3><p><strong>Step 1: Clone the Repository</strong></p><pre>git clone https://github.com/devshojol/Electron-js-boilerplate.git<br>cd Electron-js-boilerplate</pre><p><strong>Step 2: Install Dependencies</strong></p><pre>npm install</pre><p><strong>Step 3: Start in Development Mode</strong></p><pre>npm run dev</pre><p>✅ This will start:</p><ul><li>Electron main process</li><li>React frontend served by Vite</li><li>Hot Reload for instant frontend updates</li></ul><p><strong>Step 4: Build for Production</strong></p><pre>npm run build</pre><p>✅ This creates a dist/ folder with optimized frontend assets.</p><h3>📂 Project Structure</h3><pre>├── public/               # Static files (favicon, etc.)<br>├── src/<br>│   ├── assets/           # Images, SVGs<br>│   ├── components/       # Reusable React components<br>│   ├── pages/            # Pages for routing<br>│   ├── routes/           # Routes definitions<br>│   ├── App.jsx           # Main App Component<br>│   └── main.jsx          # React entry point<br>├── main/                 # Electron main process files<br>│   └── main.js<br>├── dist/                 # Production frontend output (after build)<br>├── package.json          # Project configurations and scripts<br>├── vite.config.js        # Vite configuration<br>├── tailwind.config.js    # TailwindCSS configuration<br>├── postcss.config.js     # PostCSS configuration<br>└── README.md             # Project documentation</pre><h3>📜 Available Scripts</h3><p>CommandDescriptionnpm run devRun development mode (Vite + Electron)npm run build:reactBuild the frontend (Vite)npm run buildBuild for productionnpm run lintRun ESLint to lint and format code</p><h3>🎯 Why This Boilerplate?</h3><p>When I started building apps with Electron, React, and Tailwind, I struggled every time with:</p><ul><li>Repetitive setup tasks</li><li>Broken hot reloads</li><li>Slow build tools</li><li>Unorganized project structures</li></ul><p>Now, this boilerplate solves all of that:</p><ul><li>⚡ Ultra-fast Vite development server</li><li>📦 Production-ready configuration</li><li>📜 Clean code maintained with ESLint and Prettier</li><li>🎨 Fully customizable TailwindCSS styling</li><li>📍 Router system ready to expand for multiple pages</li></ul><p>It’s designed to <strong>scale easily</strong> as your project grows.</p><h3>📜 License</h3><p>This project is licensed under the <strong>MIT License</strong>.<br> Feel free to use it, modify it, and build awesome apps on top of it!</p><h3>🔗 Get the Boilerplate</h3><blockquote><a href="https://github.com/devshojol/Electron-js-boilerplate">👉 GitHub Repository Link</a></blockquote><p>If you like it, please <strong>star</strong> the repo and share it with others who might find it useful! ⭐</p><h3>🛠 Final Thoughts</h3><p>This boilerplate is my contribution to help developers <strong>save time</strong> and <strong>focus on real innovation</strong>, not on boilerplate setup.</p><p>If you’re looking to start your next Electron desktop app, I truly believe this boilerplate can make your journey easier and faster.</p><p>Happy building! 🚀</p><p>#electron #react #vite #tailwindcss #desktopapps #opensource #programming #javascript #webdevelopment</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b84ffb9de1d8" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>