This course introduces the principles and practices of AI engineering, emphasizing software and system architecture for AI-enabled agents. Students learn how to design, build, and deploy AI agents and intelligent workflows that integrate AI models, data pipelines, APIs, and software systems. Topics include AI agent design, AI workflow orchestration, system integration, scalability, reliability, and security considerations for AI-driven systems. Through hands-on, project-based assignments, students apply AI technologies to real-world, business-driven use cases, focusing on end-to-end system design, automation, and the responsible deployment of AI solutions.
By the end of this course, students will be able to:
​
​
​
​
​
​
​
​
​
​
​
​
LO1 — Distinguish output from outcome, and defend the Person + AI Strategy. Explain why AI produces outputs while humans remain accountable for outcomes. Articulate why a Person + AI approach is essential at every stage of the AI system lifecycle — from use-case scoping to production operation.
LO2 — Apply Clean Architecture to AI-enabled systems. Treat AI models, prompt libraries, context managers, orchestration frameworks, and tool APIs as replaceable details behind clean boundaries. Design agentic systems where the Dependency Rule holds — business rules and use cases at the center, AI capabilities at the outer edges.
LO3 — Design and justify a real-world business use case for AI. Apply a Use Case Design Framework and an ROI calculator framework to select, scope, and justify a business-driven AI-enabled solution before writing code — treating the business case as the first architectural artifact.
LO4 — Apply an AI Governance & Risk Framework to a production-ready design. Address data privacy, model bias, security, cost control, human accountability, and observability. Document trade-offs through Architecture Decision Records (ADRs).
LO5 — Design, build, and deploy AI agents and workflows. Integrate AI models, data pipelines, APIs, and software systems into an end-to-end agentic workflow. Apply prompt design and context management as first-class architectural concerns — not add-ons.
LO6 — Deliver measurable results in a world of AI Agents. Present and defend an end-to-end AI-enabled system that ships measurable business value, with honest accountability for what the AI produced, what human judgment shaped, and how the system will remain safe, replaceable, and observable in production.