Anyone can call an LLM API — the hard part is building a chat experience users actually trust and enjoy. In this session, you'll see how to build a polished UI for a real AI chatbot using Blazor, based on lessons learned building a production chat assistant. We'll dig into what makes a chat UX feel responsive and trustworthy: immediacy through streamed responses and instant feedback, visibility into background processing like agent steps and tool calls, friendly conversational interactions on the primary interface, and optional access to diagnostics, metrics, and cost information for power users. You'll leave with concrete patterns — and code — for building your own AI assistant UX.

You will learn:

• How to design a chat interface that feels immediate, using streaming responses and progressive feedback in Blazor
• Patterns for surfacing background processing (agent activity, tool calls, long-running work) without overwhelming the user
• How to layer in optional diagnostics, metrics, and cost/token visibility for users who want to see what's behind the curtain