The Division of IT at Virginia Tech seeks to deploy a custom generative AI platform for use by all eligible faculty, students, and staff. The platform should leverage the existing business relationship with Microsoft and the current infrastructure investment in the Microsoft Azure cloud platform. The system should provide a chat-centric web-based user interface, direct API access for researchers and others developing custom applications, and the ability to access large language models (LLMs) trained and/or augmented using university-managed high-performance computing infrastructure. Some key features of the platform include: • Mandatory platform capabilities such as runs in existing university-managed Microsoft Azure tenant, follows Azure AI Foundry best practices, and maintains version history of product with the ability to roll back to previous versions • Security requirements such as single sign-on through Microsoft Entra ID, role-based and/or group-based access control via Grouper, and capabilities for guardrails to prevent abuse • Support requirements such as ongoing technical support and maintenance of the product, escalation path for user support issues with defined response times, and monthly or quarterly support calls with Virginia Tech administrators