Full stack JavaScript technologies & trends worth following in 2024
A look at the latest full stack JavaScript technologies that deserve attention, shaped by community interest, tech news, and market demand.
Full stack JavaScript development continues to take the software & technology world by storm and shows no sign of slowing down. In this article, I’d like to share my personal take on the trending technologies worth watching over the next few months. My picks are largely driven by what the community is focusing on and how it feels, by tech news, and by what clients are asking for. Whether you’re an experienced developer looking to sharpen your skills or a newcomer keen to chart a path in the fast-moving world of web development, I hope this post inspires you and helps you pick the right tech stack.
React 19 on the way
Nearly two years have gone by since the previous major React version came out.
Meanwhile, the well-known “create-react-app”, which bootstrapped countless React-based projects with custom tech-stacks, has been discontinued, and React’s own docs now strongly suggest choosing one of the React-powered frameworks (rather than starting from scratch and, as they put it, essentially building your own framework later). The reasoning in the docs is quite convincing, and I really encourage you to read it:
Source
React’s next major version – 19 – was announced not long ago. In my humble view, the most significant part of this release is likely the compiler, which will simplify and optimize plenty of details. For instance, with the compiler’s help, you will almost never have to worry about those awkward calls to memoization-focused hooks such as useMemo or useCallback, or functions such as memo. These have a long history of giving developers headaches, and the internet is full of examples where they were used entirely incorrectly. From now on, a compile step will handle the optimizations, so there’s no need to add them to the code by hand. It’s a win-win: the code gets optimized automatically, and it also becomes far easier to write and maintain over time. Adding a compiler is in line with trends set by competitors such as Vue or Svelte.
Other major additions are Server Actions and Server Components. Although they’ve been available and used for a while in React-based frameworks such as Next.js or Remix, React 19 is moving these features into the “stable” stage. Until now, web developers could only access them in a kind of “canary” mode, together with things like client/server directives. Overall, React appears to be heading toward an isomorphic style that blurs the boundary between server and client code. This brings a big change in how developers can reason about execution boundaries, since defining server-side actions and calling them from the client without having to hand-craft awkward API endpoints has never been this simple.
Test frameworks and tools
Testing has always been a crucial part of building software. Lately, several new and intriguing solutions have been gaining traction, and I’d like to highlight the most notable of them.
Vitest
Vitest is a fairly new test runner that comes from the same environment as Vite, which is promoted as next-generation web tooling intended to replace older tools such as Webpack. The idea is to share one project configuration between the two, so business logic and test suites, which normally go hand in hand, aren’t split apart. Built with modern web development in mind and offering a familiar API, Vitest aims to become the de facto standard for testing frontend and backend applications alike. It supports TypeScript natively, its API matches well-established community standards (think: Jest), it ships with common features such as coverage, snapshots, and mocking, and it’s simply fast thanks to running tests in parallel. Developer Experience (DX) is a key motivation behind both Vite and Vitest, and keeping unit test suites in shape with the latter is very straightforward.
Playwright
Whereas Vitest concentrates mainly on unit testing, a different framework built for end-to-end testing is gaining momentum: Playwright. It’s a cross-platform and cross-language (JavaScript/Typescript, Python, Java, .NET) solution popular with developers and automation QA engineers. Although Playwright is meant for testing web applications by automating browsers, it also provides features that can be used to test APIs. That comes from Playwright’s ability to intercept, modify and validate HTTP requests and responses, which can be used to test API endpoints directly, even if that sits outside its core focus on browser-based interactions. In earlier projects, I used playwright for exactly this and maintained several integration test suites for services I owned.
Testcontainers
While we’re on web services: they rarely stand alone and typically rely on other components such as databases, cache layers or queues. Integration testing means bringing up real infrastructure with every building block, but how do you abstract those away in unit tests? Mocking/stubbing/faking is one option (for example, swapping the underlying database client for a “fake” implementation in unit tests), but is there a better way? Meet Testcontainers: an extremely simple way to start a real instance of any service your tests need and discard it afterwards. Say your unit tests for CRUD operations should verify that everything works correctly against a real Postgres instance (rather than mocks or stubs) – there’s a module for that. Using testcontainers boils down to installing an npm package, importing and configuring the relevant module, and instantiating it in your tests. It’s well worth a try in both existing and new projects!
Runtime environments - node.js alternatives
For years, node.js was the only “serious” JavaScript runtime for backend applications outside the browser. Apart from some historical turmoil (does anyone still recall the io.js fork, now discontinued?) and tools such as Nashorn for JVM, there were practically no notable alternatives.
Deno
Deno was announced around 2018 (and version 1.0 shipped in 2020). Through Deno, Ryan Dahl, who originally created node.js, set out to fix what he saw as “broken” in node. Although it is not a “drop-in” replacement for node.js (at least not by default), it brings a number of excellent features and improvements:
- strict security and safe defaults: for instance, the developer has to grant access to the filesystem and network explicitly.
- package management: say goodbye to
node_modules! Dahl named the introduction of package.json as one of his 10 regrets, so Deno now takes another route: URLs for dependency management backed by a local cache (you can learn more about this approach here). It’s also worth noting that Deno uses the modern ES modules system by default. - first-class TypeScript support: there’s no need for extra tooling to compile, and static typing works out of the box.
Deno and node.js differ in other ways too, but these are the main selling points of the newcomer. Is it the only challenger, though?
Bun
Not at all! Bun is another runtime, promoted as a drop-in replacement for node.js (unlike Deno). Likewise, it supports TypeScript by default and internally resolves the “module madness” compatibility issues, offering interoperability between legacy CommonJS and modern ES modules with no special configuration needed. The familiar require(), import and dynamic imports (with top level await) work out of the box regardless of which module strategy the project follows.
Benchmarks show that it’s also the fastest. Bun is among my favorite and most fascinating projects around. In the words of its authors in the documentation:
“Bun is more than just a runtime. In the long run, it aims to be a unified, infrastructural toolkit for building JavaScript/TypeScript apps, with a package manager, transpiler, bundler, script runner, test runner, and more.”
It’s certainly one to watch in 2024 – what might your projects gain from Bun?
LLRT by AWS
Lastly, a quick mention of one more thing: LLRT (Low Latency Runtime), a runtime built specifically for serverless applications on AWS. It’s promoted as able to boost performance while cutting costs, but as of March 2024 it remains experimental. If your company leans heavily on AWS Lambda with JavaScript, it may be a tool to keep in mind over the next few months.
TypeScript
In the 2023 Stack Overflow Developer Survey, TypeScript again scored highly among both “most popular” and “admired/desired” technologies. Practically every new project supports static types out of the box, very often by default.
At our company, we use TypeScript for full stack development on both the frontend and the backend. We appreciate the trust & confidence it gives us in our code, along with the developer experience that type safety makes possible.
In 2024, I’m still making the case for investing in TypeScript, just as I did more than three years ago. With version 5.4 freshly released, the language continues to evolve and adoption keeps growing day by day (even though it’s already very high). Whether you’re planning an entirely new project or migrating a legacy JavaScript codebase, keeping types in check is always a sound decision. The fast development & adoption of tools and libraries such as tRPC is the best evidence of that.
Summary
I hope this thorough guide helps you stay on top of the key full stack javascript trends of 2024. If you’ve made it this far, thank you for reading! If you liked this post, feel free to share your thoughts!