Frequently Asked Questions
How does this track address the "unpredictability" of AI in standard software environments?
The track focuses on the shift from deterministic to probabilistic systems. You will learn strategies for managing the risks of non-deterministic components—such as unpredictable outputs and shifting model behaviors—by implementing strong architectural boundaries and resilience patterns that maintain system coherence.
Can I apply Domain-Driven Design (DDD) principles to AI-heavy architectures?
Yes. One of the core topics is using bounded contexts and domain models to integrate intelligent behaviors. This ensures that AI capabilities are contained within specific domains, preventing "intelligent" features from breaking the integrity of your core business logic.
What specific integration patterns will be covered for LLMs?
The sessions dive deep into RAG (Retrieval-Augmented Generation) architectures and AI microservices. You will explore patterns for designing scalable retrieval layers, model service integration, and how to manage dependencies on external AI providers within an enterprise framework.
How do we ensure governance and safety without stifling innovation?
The track covers Responsible Architecture, which involves building explicit guardrails directly into the design. This includes technical controls for safety, compliance, and traceability, ensuring that intelligent systems remain auditable and governed at the enterprise level.
Is it possible to integrate these intelligent capabilities into our existing legacy systems?
Specific focus is given to Legacy Modernization. The track provides a roadmap for introducing AI into established landscapes using integration patterns that prevent the destabilization of mission-critical, high-availability legacy systems.
