Why TypeScript

TypeScript isn’t free — it adds syntax, a build step, and a learning curve. Why do millions of developers use it anyway?

1. Bugs Caught Before Run

function send(user: { email: string }) {
  fetch(`/api/send?to=${user.email.toLowerCase()}`);
}

send({ name: "Ada" });
//   ~~~~~~~~~~~~~~~
// Argument of type '{ name: string; }' is not assignable to
// parameter of type '{ email: string; }'.

In JavaScript, this would have shipped to production — and crashed with “cannot read properties of undefined” the moment a user clicked Send.

2. Autocomplete That Knows Your Code

const user = { name: "Ada", email: "ada@x.com", admin: true };
user.
//   ↑ cursor here, editor shows: name, email, admin

Free, accurate autocomplete is the single most underrated benefit. You stop guessing field names. You stop opening other files to remember what a function returns.

3. Refactor Safely

Rename a field, change a signature, move a function — the compiler walks every call site and flags what broke. Refactors that would be unsafe in JS become routine in TS.

4. Types ARE The Documentation

function paginate<T>(items: T[], pageSize: number): T[][];

You don’t need a comment explaining the function — the signature tells you what it takes and what it returns. And the docs can’t drift out of date.

5. Confidence at Scale

For projects past ~10k lines, the compile time investment is trivial compared to the time saved on bugs and refactors.

The Costs

Realistic ones:

• Learning curve — generics, conditional types, mapped types take time
• Compile step — tsc, tsx, or a bundler
• More keystrokes — type annotations add code
• Types for libraries — npm install -D @types/foo for older libs

These costs shrink dramatically once you’re comfortable. The returns compound.

When To Use TS

The TypeScript ecosystem is now the default for serious JS — Node backends, React apps, Astro sites, library development.

Up Next

The primitives — string, number, boolean, etc.