Software Architect for AI-Driven Systems

Course overview

A software architect for AI-driven systems designs the technical structure of applications where AI components are an integral part. This training provides architectural decisions for the reliable use of AI in software, from the initial sketch to ongoing operation. Designing system architectures with AI components AI components impose different requirements on architecture than traditional components. - Define interfaces between AI services and application logic - Plan for scaling and load distribution for AI requests - Structurally consider data and model versioning - Architecturally plan for latency and response time requirements Ensuring operation and further development - Integrate monitoring for AI components into the architecture - Plan for the interchangeability of models and providers - Architecturally anchor security and data protection requirements - Structurally plan for cost control of ongoing AI use Managing risks and dependencies Architectural decisions often have long-term effects. - Critically assess dependence on individual providers - Architecturally prepare for failure behavior and emergency scenarios - Early recognition of technical debt due to rapid AI integration - Mitigate quality fluctuations of AI services in system design Designing data flows and data storage - Design data paths between applications, AI services, and storage systems - Define data quality requirements for AI components - Structurally anchor the separation of training, testing, and production data Managing collaboration across team boundaries - Document architectural decisions for development teams in a comprehensible manner - Structure collaboration with data protection and security officers - Adapt the architecture's further development to new AI possibilities - Establish architecture reviews with specialist and development teams Evaluating and communicating architecture An architecture only works if it is understood and supported. - Justify architectural decisions with evaluation criteria in a comprehensible manner - Design target images and transition architectures for gradual implementation - Prepare architecture for technical and non-technical stakeholders - Recognize and address deviations from the target architecture Deliberately balancing quality attributes - Moderate conflicts of goals between response time, cost, and quality - Realistically weigh availability requirements against effort - Treat the traceability of AI decisions as an architectural requirement - Coordinate the prioritization of quality attributes with stakeholders This training conveys how system architectures can technically integrate AI components sustainably.

Upcoming dates

• 2026-10-20T22:00:00.000Z · 2026-12-22T23:00:00.000Z · Frankfurt am Main · Hybrid Learning
• 2026-10-20T22:00:00.000Z · 2026-12-22T23:00:00.000Z · Herzogenrath · Hybrid Learning
• 2026-10-20T22:00:00.000Z · 2026-12-22T23:00:00.000Z · Leinfelden-Echterdingen · Hybrid Learning
• 2026-11-03T23:00:00.000Z · 2027-07-03T22:00:00.000Z · Virtuelles Klassenzimmer
• 2026-11-03T23:00:00.000Z · 2027-07-03T22:00:00.000Z · Frankfurt am Main · Hybrid Learning