Build AI agents that actually work — with boundaries, tools, verification, recovery, evaluation, and real-world operating discipline.
A practical professional course on designing and building a bounded AI agent that can pursue a defined goal, choose among permitted next steps, use controlled tools, observe results, verify completion, recover from failures, and escalate appropriately. The course teaches durable system-design and operating principles before platform examples. It does not recreate general workflow education, teach multi-agent orchestration, or use any single automation framework as its spine. Learners progressively assemble an AI Agent Build Portfolio and complete an integrated build-and-launch capstone. Mastery requires a score of 80% or higher.
What you will be able to do
- Classify a proposed solution as a prompt, assistant, deterministic workflow, automation, tool-using agent, or broader agentic system and defend the classification.
- Produce an agent opportunity brief that defines the goal, evidence of success, autonomy envelope, deterministic controls, prohibited actions, and conditions under which an agent should not be used.
- Design an agent information architecture that separates instructions, task context, application state, conversation history, durable memory, and retrieved knowledge, with explicit persistence and trust rules.
- Specify controlled tools using structured schemas, preconditions, permissions, confirmation gates, deterministic execution logic, result validation, and action-verification evidence.
- Construct a bounded control loop with stop conditions, retry budgets, fallbacks, human-review checkpoints, escalation paths, privacy controls, and recovery behavior.
- Create representative test cases, evaluation rubrics, failure classifications, trace requirements, security tests, and measurable release criteria for an agent system.
- Analyze and defend cost, latency, reliability, security, autonomy, and maintenance tradeoffs for a realistic agent deployment.
- Build and present an integrated bounded-agent design or working prototype specification, earn at least 80% against the capstone rubric, and package the work as an AI Agent Build Portfolio.
The AI Agentic School Decision Lab
Don’t just understand AI agents—build the system and defend the decisions. Each module ends with a realistic AI-agent build brief. You create a concrete artifact or system decision, explain your rationale, receive rigorous rubric-based coaching and a simulated stakeholder challenge, then revise until you demonstrate 80% mastery. Your best module submissions become your personal AI Agent Build Portfolio.
The curriculum is source-traceable and human-reviewed before publication. AI is used where it adds learner value—as a research tool and interactive practice/evaluation partner—not as a claimed human author, user-research participant, employer, auditor, or accreditation body.