Muneeb ur Rehman

Just launched PocketCap — a simple, clean personal finance tracker. Allocate income across custom category wallets. Track subscriptions & renewal dates. Set and hit visual savings goals. Built with Next.js & MongoDB. Try it out: pocketcap.vercel.app

Learning is great, but if you never build with what you learn, it won’t take you very far. Learning helps you understand something, but until you actually put it to work, it’s almost like you never learned it in the first place. Learn → Build → Apply → Repeat.

Knowing Next.js isn’t enough to become a better developer. After building projects, I realized I needed to understand more than frameworks. So I started learning system design to build more scalable, optimized applications. What’s one skill developers should learn beyond coding? 👇

AI has increased competition and reduced some jobs. But I love that we no longer have to spend hours memorizing syntax or searching Stack Overflow for bug fixes. Now we can focus on what matters more: building, problem-solving, and making better decisions.

I genuinely don’t understand why people dislike Gemini so much. 🤔 Is there a solid reason behind the criticism, or has it just become a trend? What’s your take?

🔥 If you’re learning to code, these 5 languages are everywhere in tech. ⚡ JavaScript → Web & Full-Stack 🐍 Python → AI/ML & Data Science 🗄️ SQL → Databases & Data ☕ Java → Enterprise & Backend 🎮 C# → .NET & Game Development Different languages. Different strengths. Which one do you use the most?

Just learned MongoDB Aggregation! 🚀 Before this, I’d fetch a user’s purchases, loop through them, and calculate the total in Node.js. Now MongoDB handles the filtering and calculations with aggregation pipelines. Small concept, but a useful addition to my backend toolkit. 💻

Web apps or desktop apps for AI? 👀 Google just launched the Gemini app for Windows. 🪟 But honestly, it doesn’t matter much to me. I use AI through web apps because switching between tabs feels easier than jumping between apps. What do you prefer: AI websites 🌐 or desktop apps? 💻

Meta just shipped Muse. 🤖 An AI agent that can send emails, book travel, make payments, and even sell a car for you. Available in the US on iOS, Android, web, and WhatsApp, powered by Muse Spark 1.3. We’re moving from chatbots to “do-bots.” What are your thoughts on this AI?

AGI era or caution era? 🤔 OpenAI calls GPT-6 “Astra” a leap toward AGI, while Jensen Huang says AGI is already here. Meanwhile, OpenAI’s top scientist warns AI may be moving faster than we can control. Are we moving too fast? Follow for more AI updates. 🤖

Which AI should you use for what? Coding & Writing Nuance → Claude Deep Multi-Step Reasoning & STEM → ChatGPT Giant Documents, Long Videos & Multimodal → Gemini Current Web Search & Sourced Answers → Perplexity Studying Your Notes & Uploads → NotebookLM Save for later. 🔖

Google just made AI music generation even more interesting. Lyria 3.5 is now available in Gemini, with more expressive vocals and richer arrangements. Choose your genre, vocal or instrumental style, and track length. AI-generated music is getting seriously good. 🎵 Try it: gemini.google.com

Nvidia is buying Hugging Face for $12.9B Hugging Face has 18M+ developers, 3M+ models, and 500K+ datasets. Nvidia already dominates AI compute and is moving deeper into the developer ecosystem. The AI race is becoming less about models alone and more about owning the infrastructure around them.

Companies have a strong incentive to make their LLMs look better on benchmarks. Sometimes, how models are trained or evaluated can make those scores look better than what you see in real-world use. A benchmark shows how well a model performs on a test, not how capable it is in the real world.

Things you stop caring about as you become a better developer: • Memorizing every bit of syntax • Chasing every new framework • Writing unnecessary code • Building everything from scratch • Making code look “impressive” **You stop trying to look good and start being one.**

What separates a good developer from a great developer? It’s not how much code you can write. It’s how well you understand problems, make decisions, find solutions, and adapt when things go wrong. Writing code is just one part of being a developer.

“Using AI will make you forget how to code.” I don’t really buy that. AI can help you write better code and explore different ways to solve a problem. You just need to understand the code—you don’t have to memorize every line. Do you think developers still need to memorize code in the AI era?

Thought I was finally done with my project. 😭 Finished the major features a week ago, but I’m still fixing UI issues, backend bugs, and random feature problems. That final 10% really takes forever. 😂 What’s your longest “I’m almost done” phase? 👀

OpenAI is ending Cursor’s direct access to its models. Following Cursor’s acquisition by SpaceX, OpenAI says it will cut off direct model access on November 12, 2026. Things are getting interesting in the AI coding space.

Building with the X API just got a little easier. Keeping track of apps, API credits, and usage could get a bit annoying. The new Developer Console puts everything in one place, so managing it is much simpler. Have you tried it yet?

Cursor is slowly becoming an all-in-one platform for developers. 👀 With its new Origin code-hosting platform, you can now create repositories, manage pull requests, browse code, and sync with GitHub—all inside Cursor.

Just completed the major features of my latest project. 🚀 Now comes the fun part: fixing all those small UI and functionality bugs. 😭 Spent the whole day fixing tiny issues, and somehow there’s still more to fix.

People think AI has made everyone a developer. But give 5 construction workers to a random person and to an experienced builder. Who’s building the better house? Same with AI. A developer knows how to use it properly. Without the fundamentals, you’ll probably end up with slop.

Happy 35th anniversary, Linux! 🐧🎉 35 years ago, Linus Torvalds shared what started as a small personal project. Today, Linux runs servers, cloud infrastructure, supercomputers, and much of the internet. 35 years later, and it’s still going strong. 🚀

Making a website responsive is a pain sometimes. 😭 Adding sm:, md:, and lg: classes for text, padding, margins, and layouts over and over gets exhausting. I faced this a lot in my project, but AI has made these repetitive tasks way less painful. Anyone else feel the same? 👀

Some of the biggest areas in software development 🌐 🥇 Web Development 🥈 Mobile Development 🥉 Data / AI / ML 4️⃣ Game Development 5️⃣ Embedded / IoT Web development is one of the largest areas, with frontend, backend, and full-stack roles covering a huge part of the industry.

Your backend shouldn’t trust the data it receives. 👀 That’s where Zod comes in. Instead of manually checking every request, Zod lets you define a schema and validate incoming data before your application touches it. Less boilerplate, fewer unexpected bugs, and much safer APIs. Do you use Zod?

When does an app actually become “vibe coded”? 👀 How much AI-generated code does a product need before you’d consider it a vibe-coded app?

A simple roadmap to becoming a web developer 🚀 1. HTML + CSS 2. JavaScript 3. Git + GitHub 4. Backend + APIs 5. Databases 6. Authentication + Security 7. React 8. Next.js 9. Testing + Deployment 10. Build real projects Don’t just learn the stack. Learn how to build, debug, secure, and ship.

What’s one coding habit you wish you had stopped earlier? 👀 For me, it was spending way too much time memorizing syntax instead of understanding the concepts behind it.

A lot of people watch tutorials and learn the concepts, but don't put them into practice. Tutorials can make you feel productive without really improving your skills. Build something, break it, debug it, and figure out why it works. That’s where the real learning happens.

VS Code offering discount on AI model 🤖

I’m about 70% done with my current finance management web app. Once I finish it, I’m planning to build a SaaS.

AI-assisted development isn’t a niche anymore. Developers using or planning to use AI: 2023 → 70% 2024 → 76% 2025 → 84% The trend speaks for itself. Source: Stack Overflow Developer Survey In what year do you think this graph will reach 100%?

I almost wrote a completely useless API call today. 😅 ​My Next.js Zustand state kept wiping on refresh, and I was about to build a new fetch route to recover it. ​Then I found persist. It auto-saves your store to localStorage so data survives refreshes instantly. Zero backend calls needed! 🚀

🚀 Cursor is becoming more than a code editor. Cursor launched Origin, bringing code hosting, repositories, and pull requests into its AI-powered workflow. The bigger shift: AI coding tools are evolving into complete development platforms, not just code editors.

5-Word AI Prompting Framework: TCRIE (Thoughtfully Create Really Excellent Input) Task: What you want AI to do. Context: Relevant background information. References: Text, files, or examples. Evaluate: Check the output against your criteria. Iterate: Refine and improve the result.

🚀 AI coding tools are becoming full development platforms. The cursor is moving into code hosting with Origin, bringing repos and PRs closer to the AI coding workflow. The future may be less about separate tools and more about one AI-native development environment.

🤖 AI coding is moving beyond “which model is best?” Different tasks need different strengths, and GitHub Copilot is embracing model choice. The future isn’t one perfect model—it’s knowing which model fits the job. The best AI developer knows when to switch.

Next.js + TypeScript + PostgreSQL + Prisma + Tailwind. The problem isn’t the stack—it’s spending weeks deciding instead of building. I’d pick a stack I’m productive with and stick with it until shipping. Less context switching, faster iteration. What’s your one-stack setup? 👀

⚔️ Imagine being stuck with ONE code editor for the next year. No switching between VS Code, Cursor, Zed, or anything else. 😭 The challenge? Picking the one that keeps your workflow fast without missing the features you rely on. I’d probably stick with VS Code. What’s your pick? 👇

🐛 The first computer “bug” was literally a bug. But the term wasn’t invented because of the moth. Engineers were already using “bug” for technical problems before 1947. The moth just became the most famous example, taped into the Harvard Mark II logbook. Did you know this? 👇

Adding one new feature shouldn't feel like rewriting half your backend. 😵‍💫 One new element can mean updating multiple models, APIs, and related entities. Good architecture makes these changes much easier to manage. How do you handle this in your projects? 👇

If you could only use 3 developer tools, which ones would survive? With hundreds of tools, extensions, and apps competing for our attention, choosing just 3 is harder than it sounds. For me, it’d be: ⚡ VS Code 🌐 Chrome DevTools 🤖 Gemini Pro What are your 3? Drop them in the comments 👇

Mobile development is becoming more seamless. Being able to control Android & iOS apps directly from VS Code means less context switching and a smoother development workflow.