Aperio Lang (aperio-lang.github.io)

A new programming language called Aperio has been introduced, aiming to bridge the gap between human cognitive models and machine execution, especially in the context of large language models (LLMs). Traditional programming languages have focused on minimizing friction during translation between human thought and machine code, but this approach creates hidden costs when utilizing LLMs. Aperio breaks this mold by employing a structural model based on recursive hypergraphs, allowing its syntax to align closely with human reasoning. This results in a more seamless coding experience, reducing the need for complex translations for each interaction with the LLM. Aperio's syntax allows developers to express their mental models in code without the additional burdens of traditional languages, such as mutex selection or error handling. For example, the code representing a matchmaker service captures the logic of queuing players and announcing matches in a way that directly corresponds to how developers conceptualize the system. The language currently supports native code generation via LLVM and is still evolving; feedback from users will help refine its semantics. This innovation is significant for the AI/ML community as it represents a shift toward programming languages that are fundamentally designed for modern computational needs, particularly within the realm of LLM-driven applications.