Jaxon + AutoSpec for GAMECHANGER

There is an urgent need to use AI to automate and greatly accelerate the lifecycle processes required to turn intelligence into actionable information within the Air Force in order to boost readiness. SAF/CN has a national Defense-related mission in the area specifically. The current best practice for problem specification creation is an entirely manual trial-and-error process done by skilled analysts. This is time consuming, expensive, noisy (in the sense of mistakes made and subtleties missed), and does not adapt well to change. The Air Force currently struggles to maintain large data sets that are well labeled, and these low quality datasets leave an opportunity to better clean, manage, and build datasets that aggregate and collect the right information. The Jaxon.AI team wants to research the best methods to solve this problem for the Air Force, and advance the development of our commercial product for use by SAF/CN. Jaxon.AI will expand the research and development effort by prototyping a solution that will build upon the current commercially-available Jaxon platform to alleviate a major manual process that remains in the ML model creation process. Jaxon's solution AutoSpec will analyze a collection of datasets (optionally with guidance around domains of interest) and propose candidate ML solution specifications. These specifications consist of detailed descriptions of ML models  as well as a graph linking the cooperative application of multiple models in order to compose a system that addresses a high-level business problem or mission objective utilizing these fine-grained models as building blocks.