Key Concepts in Architecting GenAI Applications
Generative AI has made building AI applications easier than ever. But many enterprise projects still fall short, not because of technical limitations, but because teams struggle to choose the right design patterns, development strategies, and governance practices.
This session introduces key architectural patterns for GenAI and explains how to apply them effectively in enterprise environments, while ensuring security, control, and visibility are built in from the start.
You’ll learn about:
- Core GenAI patterns: direct integrations, Retrieval-Augmented Generation (RAG), and AI agents.
- Why choosing the right GenAI architecture pattern matters, and how to decide which one fits your use case.
- The Model Context Protocol (MCP): what it is, when it’s needed, and how it standardizes how AI agents use tools.
- How an AI Gateway helps control access, enforce usage limits, and apply safety checks to GenAI APIs.
- How identity and access control can secure AI agents in scale, using zero-trust principles.
You’ll leave with practical strategies and foundational knowledge to design, build, and manage secure, governed GenAI applications at scale.
Speakers
Malith is the head of research and AI at WSO2, orchestrating the company's AI initiatives and research efforts. He is instrumental in developing AI strategy and creating AI-powered product features, fostering a culture of innovation among the AI team for seamless project progression from inception to delivery. Malith also engages in collaborative research with academic institutions, aiding the advancement of computer science. He is a regular speaker at prominent developer events such as DeveloperWeek, the Global Big Data Conference, and DEV DAY, sharing insights and promoting knowledge exchange within the tech community. Malith has made significant academic contributions, publishing his work in journals like IEEE Transactions on Parallel and Distributed Systems (TPDS) and the Journal of Parallel and Distributed Computing (JPDC), and presenting at conferences such as IEEE Cluster, IEEE NCA, and DEBS. Malith holds a PhD in computer science from RMIT University in Australia.
Nadheesh Jihan is an AI leader at WSO2, driving AI adoption across WSO2 products. With expertise in generative AI, deep learning, and machine learning, he has been instrumental in shaping AI strategy and delivering enterprise-grade solutions. Beyond AI, his work spans system performance, distributed systems, and API management. Nadheesh has played a hands-on role in integrating AI into real-world use cases, including enhancing WSO2 Choreo with AI-driven capabilities. He also contributes to enabling middleware products for AI development, helping enterprises build and scale production-grade AI solutions. Passionate about AI’s impact on enterprise software, he continues to bridge innovation with practical applications.