We will present a modular, multi-agent coding assistant designed to streamline and enhance the coding experience for clinical insights generation using R packages. Built on a collaborative multi-agent architecture, the system is composed of specialized agents that handle distinct tasks. It intelligently blends publicly available resources with Roche’s internal process knowledge to deliver context-aware code suggestions. An embedded webR environment enables rapid, in-app execution for immediate feedback and iteration. We are currently exploring integration via MCPs to extend the assistant’s utility into IDEs such as VSCode and Positron, ensuring seamless support across the developer workflow.