Exploring AI at Red Hat Summit 2025: Meet Hope, the AI Avatar

Key Concepts

• AI Avatar (Hope): An AI application representing CDW, designed to interact with users.
• OpenShift: A containerization platform used to host and manage Hope's application components.
• On-Premise Data Center: The physical location where Hope's backend services are hosted.
• OpenShift Router: A feature used to expose Hope to the internet.
• Front End/Back End Architecture: Separation of the user interface (Hope's avatar) from the data processing and logic.
• API (Application Programming Interface): The interface through which the front end and back end communicate.
• LLM (Large Language Model): A powerful AI model used for natural language processing.
• Azure Speech to Text: A cloud-based service used to convert speech to text and vice versa.
• Availability: Ensuring Hope remains operational even in the event of outages or disasters.
• Authentication: Verifying the identity of users accessing Hope.

1. Introduction to Hope, the AI Avatar

• Tim Way, a Senior Consulting Engineer at CDW, introduces Hope, an AI avatar.
• Hope is described as "just another application on the network."
• The discussion focuses on the architecture and infrastructure supporting Hope.

2. Infrastructure and Deployment

• Hope runs on OpenShift in CDW's on-premise data center.
• The application consists of multiple services running within the OpenShift framework.
• OpenShift Router is used to make Hope accessible from the internet.
• All internet traffic is routed through OpenShift to the application stack.

3. Security and Availability

• Security is a concern for both the front end and back end of the application.
• Authentication is implemented at both the application level and the infrastructure level.
• F5 products can be used to provide an additional security layer in front of the application.
• Availability is a key consideration, with plans to make Hope globally available.
• Disaster recovery and business continuity are important factors in ensuring availability.

4. Front End and Back End Communication

• Hope's front end communicates with the back end via an API.
• The API is exposed so that other front ends (e.g., different avatars) could be connected to the same back end.
• The front end handles the user interface and presentation, while the back end handles data processing and logic.

5. LLM and Speech Processing

• Hope uses a combination of local and remote processing for language understanding and generation.
• A local model runs on the display system, while a separate system in the back end handles more intensive processing.
• Azure Speech to Text is used to convert speech to text and vice versa.
• The process involves converting speech to text via the Azure API, sending the text to the back end, receiving a text response, and converting the text back to speech.
• The speed and efficiency of this process are noted, considering the network latency and token processing involved.

6. Future Plans and Applications

• Future plans for Hope include feeding her with more information about CDW's consulting services.
• Improving Hope's availability is a key technical goal.
• CDW sees Hope as a tool for discussing AI applications, availability, and their services in the OpenShift and F5 space.
• Potential future versions of Hope could be customized to represent different individuals or roles.

7. Notable Quotes

• Tim Way: "So Hope is our AI avatar. What's interesting to me about Hope is she's just another application on the network."
• Tim Way: "So for Hope, for me, is once the application code is written, we're looking for a place to run it. Okay. So that I can get on the network and start to, to receive requests."

8. Technical Terms

• AI Avatar: A digital representation of a person or entity powered by artificial intelligence.
• OpenShift: A container platform that provides a way to manage and deploy applications in containers.
• On-Premise Data Center: A data center located within an organization's physical premises.
• OpenShift Router: A component of OpenShift that routes external traffic to the appropriate services within the cluster.
• API (Application Programming Interface): A set of rules and specifications that software programs can follow to communicate with each other.
• LLM (Large Language Model): A type of AI model trained on a massive amount of text data, capable of generating human-like text.
• Azure Speech to Text: A cloud-based service provided by Microsoft Azure that converts spoken audio into written text.

9. Logical Connections

• The video begins by introducing Hope and then delves into the technical architecture that supports her.
• The discussion moves from the on-premise deployment to considerations of security and availability.
• The conversation then explores the front end/back end communication and the role of LLMs and speech processing.
• Finally, the video concludes with a look at future plans and applications for Hope.

10. Synthesis/Conclusion

Hope, the AI avatar, represents a practical application of AI technology within CDW. The architecture involves a multi-layered approach, leveraging OpenShift for containerization, Azure for speech processing, and a combination of local and remote processing for language understanding. Key considerations include security, availability, and the ability to integrate with various front ends. Future plans focus on enhancing Hope's knowledge base and improving her overall availability, positioning her as a valuable tool for showcasing CDW's services and expertise in the AI and cloud space.