Why a $5 Million, Five‑Year Language Project Ended Up Switching to TypeScript

Technical founders often face a core choice: evolve within an existing ecosystem or build a brand‑new system from scratch. The latter can appear more disruptive and attract media attention, but it also carries significant risk.

Wasp was founded in 2020 with the ambition to become the "Rails or Laravel of the JavaScript world," offering a full‑stack framework that abstracts common web‑app patterns such as authentication, email, and deployment configuration. After receiving Y Combinator backing in 2021, the team raised over five million dollars and released a custom domain‑specific language called wasp‑lang.

A representative snippet of the final wasp‑lang syntax shows how an application, its routes, authentication, and queries are declared declaratively:

app todoApp {
  title: "ToDo App", // visible in the browser tab
  auth: { // full‑stack auth out‑of‑the‑box
    userEntity: User,
    methods: { google: {}, gitHub: {}, email: {...} }
  }
}

route RootRoute { path: "/", to: MainPage }
page MainPage {
  authRequired: true, // Limit access to logged in users.
  component: import Main from "@client/Main" // <-- Your React code.
}

query getTasks {
  fn: import { getTasks } from "@server/tasks", // <-- Your Node.js code.
  entities: [Task] // Automatic cache invalidation.
}

Once the product reached users, the team observed a sharp drop in enthusiasm. Although developers liked the idea of abstracting repetitive patterns, the requirement to learn a new language caused immediate hesitation. The "lang" suffix in the language name triggered a reflexive alarm: professionals already juggle many languages and are reluctant to add another.

Beyond the language itself, the deeper challenge lay in the toolchain. Building comprehensive IDE support for a custom language proved extremely resource‑intensive. The VS Code extension achieved only about 80 % of the team's expectations, and essential tooling such as debuggers, static analysis, code formatters, and a rich ecosystem of third‑party libraries remained out of reach for a small startup.

After years of practice, Sosic distilled the core insight: "A language is never a moat; the real value is compile‑time, high‑level understanding of the whole application." The same capabilities—automatic code generation, dependency analysis, and full‑stack abstraction—can be realized on top of the existing TypeScript ecosystem without a bespoke language.

Consequently, the Wasp team decided to replace the custom DSL with a TypeScript SDK. The underlying architecture, compiler, and code‑generation logic stayed unchanged; developers now write familiar TypeScript, gaining instant IDE support, type checking, and auto‑completion while still accessing Wasp's full‑stack abstractions.

This shift dramatically lowered the adoption barrier: developers no longer need to learn a new language, only to be proficient in TypeScript, which is already a baseline skill for modern web developers. The trade‑off is writing slightly more code, but the benefit is full compatibility with existing editors, debuggers, NPM packages, and code‑style tools—making the proposition far more attractive to most teams.

The broader lesson highlighted by the article is that in domains where the toolchain is already mature, introducing a new language carries high learning‑costs that must be justified by substantial product value. Established languages bring decades of compiler optimizations, extensive libraries, and community support that are difficult for a startup to replicate quickly.