Write code in Tamil, Tanglish, or English. TEN compiles every dialect to one canonical AST, then down to native machine code through LLVM — with Rust-inspired safety and a grammar designed for reliable AI code generation.
TEN separates syntax from vocabulary. The same keyword resolves to one canonical token whether you write let, mathippu, or மதிப்பு — so the compiler, your tools, and AI models all reason about one language underneath many surfaces.
Native keywords in Tamil, Tanglish, and English, with more Indian languages on the roadmap.
Predictable, unambiguous grammar designed for accurate generation and static analysis by AI models.
Compiles through LLVM straight to machine code — no interpreter, no VM overhead.
Types are checked at compile time with clear, actionable diagnostics.
One toolchain targeting Windows, Linux, macOS, and beyond.
Immutable-by-default design, with a borrow checker planned for v1.0.
Every dialect compiles to the exact same AST. Switch tabs — the logic never changes, only the words used to express it.
let age = 20 if age > 18 print("Adult") end
mathippu vayathu = 20 endraal vayathu > 18 sol("Adult") mudivu
மதிப்பு வயது = 20 வயது > 18 என்றால் சொல்("Adult") முடிவு