Cloud computing is essential to a growing number of science use cases, but configuring scientific environments for deployment in the cloud can be challenging given that it often requires specialized knowledge in system administration, networking, security, and involves numerous configuration settings. Furthermore, many scientific workloads depend on tightly coupled virtual clusters, specialized hardware, fast interconnects, accelerators, and custom drivers. Managing this complexity consumes valuable time and attention that researchers could otherwise devote to science. This project develops an AI-based conversational assistant for configuring scientific computing environments that lets researchers describe what they need in everyday language and then creates the environment and verifies its integrity on shared research cloud infrastructure. The benefits of this approach range from increasing scientific productivity and lowering the cost of using cloud computing to enabling practical reproducibility of computational experimentation. The project designs and deploys an AI-based agent framework that can plan, provision, and validate scientific computing environments on open research computing infrastructure, such as Chameleon and Jetstream2. The framework combines large language models running on open, high-performance academic hardware with a set of software tools exposed through standard interfaces that include cloud-based services for resource provisioning, hardware and so