AI-Powered Solutions with CDW's Jonathan Hooker at Red Hat Summit 2025

Key Concepts

• Hybrid Cloud Architecture: Combining on-premises infrastructure with cloud services.
• OpenShift: A container platform for deploying and managing applications.
• LLM (Large Language Model): An AI model trained on vast amounts of text data.
• Agentic LLM: An LLM equipped with tools to interact with external systems and data sources.
• REST API: An interface that allows different software systems to communicate.
• Role-Based Access Control (RBAC): A security mechanism that restricts system access to authorized users.
• Data Governance: Policies and procedures for managing data quality, security, and access.

Hope: A Digital Avatar for CDW

Introduction

Jonathan Hooker, a Cloud Architect with CDW, introduces Hope, a digital avatar designed to explain CDW's services and capabilities. Hope is deployed at various conferences, including Red Hat Summit and Cisco Live. The goal is to demonstrate practical AI applications for businesses beyond simple chatbots.

Architecture of Hope

• Enterprise Solution: Hope is not just a front-end avatar but an overall enterprise solution architecture.
• Cost Efficiency: The architecture aims to share expensive AI hardware (NVIDIA or Intel) across multiple use cases.
• Backend (OpenShift): The LLM resides on OpenShift AI and is fully agentic, with tools to search websites and interact with uploaded documents.
• REST API: A REST API acts as an interface layer, allowing multiple front-end implementations (avatars, chat windows, custom applications) to interact with the backend.
• Scalability: The separation of front-end and back-end components enables scalable deployment across various use cases.

Data Security

• API Authentication: The API layer enforces authentication to control data access for the LLM.
• Role-Based Access Control: Authentication controls which data sources the LLM can access, preventing unauthorized access to sensitive information (e.g., financial data).
• Data Protection: The architecture is designed to protect data throughout the AI process, ensuring that only authorized personnel can access specific data sets.

Data Management

• Use Case-Driven Approach: Data organization is based on specific use cases.
• Tool-Based Organization: Common data types are grouped together into "tools" with descriptions for the LLM.
• Example: Financial Excel documents and marketing Word documents are stored separately, allowing the LLM to differentiate between data types and access the appropriate tool.
• Contextual Relevance: The system recognizes the need for precision in financial data versus creativity in marketing data.

Scalability and Security

• Scalability: The architecture is designed to be scalable and deliver solutions globally.
• Security: The security layer prevents unauthorized access to sensitive data, ensuring that financial data is not accessible to marketing personnel or interns.
• API Control: The LLM only accesses tools for which the user is authenticated.

Roles and Responsibilities

• IT and Data Teams: Manage the backend, control data access, and implement role-based access control.
• Developers: Access the API to build front-end applications like avatars or chat interfaces.

Data Quality

• Careful Planning: Data ingestion requires careful planning and interaction with data teams.
• Data Selection: CDW's data resources help clients select and ingest data effectively.
• AI Quality: The quality of AI output depends on the quality of the input data.
• Data Governance: Avoid simply dumping folders of data without proper selection and curation.

F5 Integration

• F5 solutions are integrated to enhance security and scalability.

Live Demo of Hope

• Hope provides information about CDW, including its role as a technology solutions provider, its range of services (IT consulting, cloud solutions, cybersecurity), and its goal to drive business growth and improve efficiency for clients.

Conclusion

Hope demonstrates a practical application of AI that delivers business value by providing a scalable, secure, and data-driven solution for customer engagement and information delivery. The architecture emphasizes data governance, security, and the separation of front-end and back-end components for flexibility and scalability.