The rise of Generative AI and large language models (LLMs) is transforming how we design, develop, and interact with modern software systems. In this session, we’ll explore how to architect systems that seamlessly integrate AI capabilities into both the front end and back end, paving the way for more intelligent, responsive, and adaptive applications.

We’ll examine how Generative AI changes the user experience with smart forms, text-to-speech interfaces, and avatar integration. You'll learn how LLMs can be leveraged in the backend to replace traditional algorithms, offering new ways to process data and handle complex tasks. We’ll also discuss strategies for mitigating the inherent fuzziness of LLM results to maintain system reliability.

The session will cover the use of frameworks and libraries like Semantic Kernel to bridge the gap between classical programming paradigms and AI-driven logic. We’ll explore using English as the communication contract between services, moving beyond traditional API contracts like JSON/REST, GraphQL, or gRPC. Imagine a world where services communicate by requesting information or actions in plain English—what new possibilities and challenges does this unlock?

Security implications, cost considerations, responsiveness, and other quality attributes will also be addressed. Attendees will gain a comprehensive understanding of how to design and build systems ready to harness the full potential of Generative AI.