Systems | Development | Analytics | API | Testing

In this episode, *Dr. Sanjiva Weerawarana* and *Asanka Abeysinghe* demystify what “agents” really are and why architects should care. They walk through core concepts and terminology—agents, agent loops, prompts, context, memory, RAG, tools, MCP, and skills—and discuss how agents observe, act, and evaluate. The conversation compares agents to traditional systems, explores where agents fit in modern architectures (including solo agents, agent-to-agent patterns, and multi-agent setups), and looks at orchestration challenges.

In this video, we dive into the new MCP Gateway features in WSO2 API Manager 4.6.0. Learn how to onboard and manage existing Model Context Protocol (MCP) servers in minutes.

Prior authorization has long been one of healthcare’s most time-consuming and costly administrative processes, creating friction between payers, providers, and patients. Manual data entry, back-and-forth faxes, and complex coverage rules often delay care delivery and frustrate both clinicians and members. In this webinar, we'll explore how conversational AI — combined with FHIR-based interoperability, contextual understanding, and an agentic approach — can revolutionize the prior authorization process. Discover how AI-driven, human-like interactions can transform prior authorization into a seamless, intelligent workflow.

“Agentic” is the defining word of 2026. While Large Language Models (LLMs) serve as the brain , AI Agents are the limbs —entities that take action, interact with real systems, and make autonomous decisions. In this deep-dive session from the WSO2 2026 Technology Conference, Ayesha Dissanayaka from the WSO2 Identity & Access Management (IAM) team demystifies what truly makes an AI agent—and tackles the most critical enterprise challenge.

Architecture diagrams are crucial, but the complexity of modern systems demands a better way to communicate. WSO2 CTO *Asanka Abeysinghe* and Lead Solutions Engineer *Erandi Ganepola* break down the L0-Ln Methodology—a simplified, level-based approach to visual abstraction. Watch our Words Unplugged video to discover how this scales from a team view to an organization-wide enterprise architecture, providing durable anchors (L0) and change-friendly details (L1-Ln) that evolve with the business.

AI agents are becoming more autonomous — making decisions, taking actions, and accessing sensitive data — but traditional IAM systems weren’t built for them. How do you secure, control, and audit AI agents in your apps? In this video, Savindi Waduge introduces *WSO2 Identity Server 7.2,* a major release designed to address the security challenges of the AI era. From defining "Identity for AI Agents" to securing the *Model Context Protocol (MCP)* , this update ensures your autonomous systems are secure, accountable, and manageable.

Even AI experts like Andrej Karpathy don’t let AI do the thinking for them. The rule? Manual First. If you let an algorithm do the first pass, you lose the nuance and "sense" that only a human brain can pick up. Don’t trade deep understanding for a shortcut. In this clip from EP18: Deep Work for Architects, Dr. Sanjiva Weerawarana explains why relying on AI summaries kills your ability to think deeply—and how to use tools without losing your edge.

Slides kill real conversations. Here’s what great architects do instead…