Rex - Parallel functional language for scientific workflows
Show HN: Rex, a parallel functional language for scientific workflows
Rex is a statically typed, pure functional workflow language designed for scientific computing and data processing. It combines a real functional language with content-addressable storage, typed tool APIs, and isolated Docker execution to make workflows concise, inspectable, and parallelizable. Rex treats data as immutable values, uses BLAKE3 hashes for artifact identity, and provides typed interfaces to tools like FFmpeg and ImageMagick, ensuring safe and reproducible pipelines. It also offers a clean boundary for LLM-generated workflows, with static types providing fast feedback and a closed tool boundary limiting execution risks.
Rex starts with a small general-purpose language instead of a DAG, making control flow, data flow, reuse, and error handling part of one expressive language rather than a mixture of YAML, shell, and application-specific configuration.