Communities generate valuable insights every day through conversations across platforms such as WhatsApp, Meetup, LinkedIn and other social channels. Hidden within these discussions are patterns that reveal community health, member engagement, learning interests, support needs, emerging topics and opportunities for stronger collaboration. However, bringing these signals together and analysing them consistently is difficult and time-consuming. RWiT Compass is an AI-powered Community Intelligence Platform designed to transform these distributed conversations into structured, measurable insights. It combines data engineering, AI and business intelligence to convert unstructured community interactions into analytics that help community leaders make more informed decisions. The current MVP focuses on exported WhatsApp conversations as the first data source. It parses chat histories, anonymises member identities, reconstructs conversations, extracts AI-generated signals and produces Power BI-ready datasets and dashboards. The platform is designed with a modular architecture that will progressively integrate additional community platforms such as Meetup, LinkedIn and Slack, providing a unified view of the entire community ecosystem. Rather than relying on assumptions or occasional surveys, RWiT Compass enables organisations to understand the overall health of their communities through measurable indicators, including participation trends, activity levels, discussion topics, sentiment, learning interests, support requests and emerging opportunities. This presentation will demonstrate how AI, Python and Power BI can transform fragmented community data into actionable community intelligence while maintaining privacy through anonymisation and responsible AI practices.