Show HN: Moss, an AI-led programming language experiment (github.com)

Moss is an innovative experimental programming language that facilitates collaboration between humans and AI agents over long-lived software projects. Developed with the help of OpenAI's Codex and community contributor Fujo930, Moss is a public repository showcasing its AI-led design, implementation, and documentation process. Currently at version 0.2.0, Moss features a compact interpreter capable of executing useful example programs and has begun to support self-hosting functions, although it is still considered alpha software. This project is significant for the AI/ML community as it explores the potential of AI in software development, raising important questions about collaboration between human developers and AI systems. Moss includes key technical features such as a lexer, parser, type declarations, and runtime type contracts, all of which are integral to its functioning. The repository outlines various commands for running, checking, and testing code within the Moss ecosystem, and it aims to enhance AI understanding of programming paradigms while fostering a new level of sophistication in language design. This synthesis of AI and programming brings fresh perspectives on the future of software engineering.