Weather Machine Learning Platform Research to Operations (WxMLP R2O)

NextGen Federal Systems (NextGen) seeks to integrate and operationalize existing capabilities and their SBIR results into a complete and operational Weather Machine Learning Platform (WxMLP). The proposed WxMLP Research to Operations effort directly addresses the Air Force Science and Technology Strategic Capability of Rapid, Effective Decision-Making. This Strategic Capability aims to increase the speed of battlespace understanding and decision-making and enable us to act faster than any adversary. It states, “Artificial intelligence and predictive analytics are essential,” and the Air Force must overcome “unpredictable and uncertain physical environments, noisy and unstructured data from dissimilar sources, [and] limited training data for machine learning.” To achieve this vision, NextGen proposes to build and deploy a coherent, comprehensive, universal, operational application. WxMLP leveraged the Air Force Research Laboratory (AFRL) Streamlined ML ecosystem as a foundation for operationalizing ML, enhanced it, and extended it to work with spatiotemporal data to meet AFW needs.  NextGen will work with the AFWE Ops Team to coordinate employment of the AFWE DevSecOps toolchain and CI/CD pipeline to get WxMLP securely running in the AFWE CP and continually improving capability over its entire life cycle. Together, the robust, resilient, cyber-secure, easy-to-use, certified, operational WxMLP application will empower AF Weather personnel to affordably leverage the amazing breakthroughs in ML application to meet many of their currently unmet mission needs.