Rust produces software that is fast, memory-safe and hard to break: whole classes of bugs are caught by the compiler instead of in production. It is also cheap to run: a small memory and CPU footprint means smaller machines and lower bills. It is our main language for new systems.
Where we use it
- Backend services and APIs that must stay fast and predictable under load.
- Data processing: collecting, validating and transforming large volumes of data.
- Rust inside Python: speed-critical parts written in Rust and used from Python through PyO3, so data and analytics teams keep their tools.
- Command-line tools and libraries that ship as a single binary.
Not everything needs Rust
Python, Java or Kotlin are often the better fit, for example when a team or an existing ecosystem already works in them. We choose the language per problem and keep the interfaces clean, so parts can move later.