Expanding Multilingual Interoperability in the COGNIT Device Runtime

COGNIT is an AI-enabled Adaptive Serverless Framework for the Cognitive Cloud-Edge Continuum that enables the seamless, transparent and trustworthy integration of data processing resources across public providers and on-premises data centres. The Device Runtime is the on-device library that abstracts authentication, scheduling, Edge Cluster discovery, latency measurement and function offloading. It is currently distributed as two reference implementations: device-runtime-py (Python, v4.0, Nov 2025) and device-runtime-c (C, v4.0, Dec 2025), both released under Apache 2.0. This white paper proposes the next evolutionary step of the framework: opening the Device Runtime to additional programming languages such as Rust, Go, Java/Kotlin, JavaScript/TypeScript and C++. Building directly on the source code and public wikis of both reference implementations, the document explains the real architecture currently in production, identifies the language-agnostic boundaries (REST endpoints, JSON envelope, Biscuit-token authentication, 16 KB function payload limit), and outlines integration approaches, architecture considerations, security and footprint constraints, and a set of best practices for delivering production-grade multi-language interoperability.