Cloud computing is a well-studied topic, but it does face some challenges, such as a lack of dynamic service deployment and selection features that can support both edge and cloud computing environments. When end users and service providers have no choice but to deploy their services in the cloud, performance issues for latency constrained cloud ap-plications may arise. To address such challenges, we proposed ClouEdge (i.e., Cloud-Edge), an optimized cloud computing infrastructure built on a JYAGUCHI computing framework that supports categorization of the services before the deployment is executed. It ensures secure and dynamic service delivery and provides deployment options to the users. End users can decide the location of deployment either in cloud or edge as per the sizes of the services that work in both cloud and edge computing environments. The pro-posed architecture is a novel system built on top of a public JYAGUCHI platform that can dynamically optimize and deploy services in the cloud or on-premises based upon user intention and requirement. We argue that in order to attain the appropriate level of quality, additional processes and operation layers are needed in addition to just delivering cloud resources to the edge. This architecture also performed well for latency-constrained cloud applications. We evaluated our system by testing the resource efficiency of each compo-nent. Applications and services that are latency-sensitive, will benefit from our proposed architecture such as various distributed IoT and AI services, which do not fit well with the new concepts and platforms available today.