← Back to posts
Apr 09, 2026
The Stack I'd Learn in 2026 If I Were Starting Over
Opinionated, practical, and based on what actually gets you hired and paid. Not what's trending on Twitter.
Hi Friends,
What’s Inside The exact stack I’d choose if I were starting from zero this year, why I’d choose it, and what I’d skip entirely. No hype, no “it depends,” just a real opinion from someone actively building, freelancing, and shipping.
Why You Should Subscribe I write from the intersection of building products, freelancing, and learning in public. If you want opinions from someone who’s actually in the trenches and not a content creator recycling the same advice, you’re in the right place.
Yesterday I told you I had strong opinions about this. So here they are.
Everyone has a “what to learn in 2026” post. Most of those posts come from people who haven’t shipped a product in years, or from influencers whose job is to make content about code, not write it.
I’m building a mobile app. I’m freelancing for clients through FabBuilds. I’m studying computer science. I’m in it every single day. So here’s what I’d actually learn if I wiped the slate clean and started over in 2026.
The Language: JavaScript/TypeScript
Not because it’s the best language. It’s not. But it’s the most versatile. You can build frontends, backends, mobile apps, serverless functions, and CLIs with one language. When you’re starting out, reducing context switching is everything.
TypeScript specifically because the industry has moved there and it’s no longer optional. Every serious codebase you’ll encounter in a job or freelance project expects it.
I’d skip Python initially. I know that’s controversial. Python is great and I use it. But if your goal is to get hired or land freelance work building products people use, JavaScript gets you there faster because it covers more surface area.
The Frontend: React and Next.js
React isn’t going anywhere. The ecosystem, the job market, the tooling, the community. It’s enormous. Love it or hate it, it’s the standard.
Next.js on top of React because it handles routing, server-side rendering, API routes, and deployment in one framework. You’re not stitching together five different tools. You open a Next.js project and you have a production-ready setup immediately.
I’d skip Vue and Svelte. Not because they’re bad. They’re lovely. But the job market is smaller and the freelance demand isn’t there at the same scale. When you’re starting out, optimize for opportunity, not elegance.
The Mobile Framework: React Native with Expo
If you already know React, React Native is basically free. The mental model is the same. The component structure is the same. You’re writing the same language.
Expo specifically because it eliminates the native toolchain headaches that kill motivation for new developers. You don’t need Xcode configured perfectly. You don’t need Android Studio. You run npx expo start and your app is on your phone.
I’d skip Flutter. Dart is a fine language but it’s another thing to learn, another ecosystem to navigate, and the crossover to web development is minimal. React Native lets you share knowledge between web and mobile. Flutter doesn’t.
This is the controversial one.
PHP gets made fun of constantly. And Laravel is the thing that makes PHP not just tolerable but genuinely excellent. It’s the most complete backend framework I’ve used. Authentication, database migrations, queues, scheduling, mail, file storage, API resources. It’s all built in, documented beautifully, and works out of the box.
The freelance market for Laravel is massive. Small businesses, agencies, startups, SaaS products. There’s real money in Laravel and significantly less competition than the JavaScript-everything crowd.
I’d learn Node.js too eventually, but I wouldn’t start there. Express is too barebones. You end up making fifty decisions before you’ve written a single feature. Laravel lets you focus on building.
Let me know your thoughts on this one because there is a big debate on this!
Leave a comment
The Database: MySQL
Not Postgres. Not MongoDB. MySQL.
Because Laravel’s default database is MySQL, most tutorials and packages assume it, and for the type of projects you’ll build starting out, it does everything you need. Postgres is fantastic and I’d learn it later. But MySQL gets you moving with the least friction.
I’d skip MongoDB entirely unless a specific project requires it. Most beginners reach for it because “NoSQL sounds modern.” For 90% of applications, a relational database is the right choice and fighting that early causes more confusion than it solves.
What I’d Skip Entirely
Rust. Incredible language. Not for beginners. Not for most product work. Learn it in year three if systems programming interests you.
GraphQL. Adds complexity that most projects don’t need. REST is simpler, better understood, and totally sufficient until you’re at a scale where GraphQL actually pays off.
Kubernetes. You don’t need container orchestration to ship a product. Vercel, Railway, Laravel Cloud. These exist so you don’t have to think about infrastructure on day one.
Web3/blockchain. The job market contracted. The hype cycle moved on. Unless you’re specifically passionate about it, your time is better spent elsewhere.
The Real Answer
Remember what I said yesterday? The stack matters less than people think. What matters is picking one, going deep, and building real things with it. The developers who struggle aren’t the ones who picked the “wrong” framework. They’re the ones who spent a year evaluating frameworks and never built anything.
Pick a stack. Build a project. Ship it. Get feedback. Improve it. That loop is worth more than any amount of research.
This is the stack I’d use to run that loop as fast as possible.
Let’s Build It Beautifully,
Fab
What’s Inside The exact stack I’d choose if I were starting from zero this year, why I’d choose it, and what I’d skip entirely. No hype, no “it depends,” just a real opinion from someone actively building, freelancing, and shipping.
Why You Should Subscribe I write from the intersection of building products, freelancing, and learning in public. If you want opinions from someone who’s actually in the trenches and not a content creator recycling the same advice, you’re in the right place.
Yesterday I told you I had strong opinions about this. So here they are.
Everyone has a “what to learn in 2026” post. Most of those posts come from people who haven’t shipped a product in years, or from influencers whose job is to make content about code, not write it.
I’m building a mobile app. I’m freelancing for clients through FabBuilds. I’m studying computer science. I’m in it every single day. So here’s what I’d actually learn if I wiped the slate clean and started over in 2026.
The Language: JavaScript/TypeScript
Not because it’s the best language. It’s not. But it’s the most versatile. You can build frontends, backends, mobile apps, serverless functions, and CLIs with one language. When you’re starting out, reducing context switching is everything.
TypeScript specifically because the industry has moved there and it’s no longer optional. Every serious codebase you’ll encounter in a job or freelance project expects it.
I’d skip Python initially. I know that’s controversial. Python is great and I use it. But if your goal is to get hired or land freelance work building products people use, JavaScript gets you there faster because it covers more surface area.
The Frontend: React and Next.js
React isn’t going anywhere. The ecosystem, the job market, the tooling, the community. It’s enormous. Love it or hate it, it’s the standard.
Next.js on top of React because it handles routing, server-side rendering, API routes, and deployment in one framework. You’re not stitching together five different tools. You open a Next.js project and you have a production-ready setup immediately.
I’d skip Vue and Svelte. Not because they’re bad. They’re lovely. But the job market is smaller and the freelance demand isn’t there at the same scale. When you’re starting out, optimize for opportunity, not elegance.
The Mobile Framework: React Native with Expo
If you already know React, React Native is basically free. The mental model is the same. The component structure is the same. You’re writing the same language.
Expo specifically because it eliminates the native toolchain headaches that kill motivation for new developers. You don’t need Xcode configured perfectly. You don’t need Android Studio. You run npx expo start and your app is on your phone.
I’d skip Flutter. Dart is a fine language but it’s another thing to learn, another ecosystem to navigate, and the crossover to web development is minimal. React Native lets you share knowledge between web and mobile. Flutter doesn’t.
This is the controversial one.
PHP gets made fun of constantly. And Laravel is the thing that makes PHP not just tolerable but genuinely excellent. It’s the most complete backend framework I’ve used. Authentication, database migrations, queues, scheduling, mail, file storage, API resources. It’s all built in, documented beautifully, and works out of the box.
The freelance market for Laravel is massive. Small businesses, agencies, startups, SaaS products. There’s real money in Laravel and significantly less competition than the JavaScript-everything crowd.
I’d learn Node.js too eventually, but I wouldn’t start there. Express is too barebones. You end up making fifty decisions before you’ve written a single feature. Laravel lets you focus on building.
Let me know your thoughts on this one because there is a big debate on this!
Leave a comment
The Database: MySQL
Not Postgres. Not MongoDB. MySQL.
Because Laravel’s default database is MySQL, most tutorials and packages assume it, and for the type of projects you’ll build starting out, it does everything you need. Postgres is fantastic and I’d learn it later. But MySQL gets you moving with the least friction.
I’d skip MongoDB entirely unless a specific project requires it. Most beginners reach for it because “NoSQL sounds modern.” For 90% of applications, a relational database is the right choice and fighting that early causes more confusion than it solves.
What I’d Skip Entirely
Rust. Incredible language. Not for beginners. Not for most product work. Learn it in year three if systems programming interests you.
GraphQL. Adds complexity that most projects don’t need. REST is simpler, better understood, and totally sufficient until you’re at a scale where GraphQL actually pays off.
Kubernetes. You don’t need container orchestration to ship a product. Vercel, Railway, Laravel Cloud. These exist so you don’t have to think about infrastructure on day one.
Web3/blockchain. The job market contracted. The hype cycle moved on. Unless you’re specifically passionate about it, your time is better spent elsewhere.
The Real Answer
Remember what I said yesterday? The stack matters less than people think. What matters is picking one, going deep, and building real things with it. The developers who struggle aren’t the ones who picked the “wrong” framework. They’re the ones who spent a year evaluating frameworks and never built anything.
Pick a stack. Build a project. Ship it. Get feedback. Improve it. That loop is worth more than any amount of research.
This is the stack I’d use to run that loop as fast as possible.
Let’s Build It Beautifully,
Fab