Agentic AI System Design / Reference
The reference library
The artifacts this course keeps coming back to. Every template here is the one a design review, a security review, or an on-call rotation would actually ask you for, stripped to the fields that change a decision.
Method
The design process the whole course applies.
Canvas
The worksheet you fill for every system.
Template
Fill-in artifacts a real design review expects to see.
- Tool contract template The typed, risk-rated contract every tool needs before it ships.
- Threat model template Assets, threats, boundaries, prevention, detection.
- Eval plan template What to measure, on which data, with which release gate, across components, trajectories, and outcomes.
- SLO and incident runbook Indicators, objectives, alerts, and the runbook an operator uses at 3am.
- Architecture decision record One decision, its options, the choice, and what it costs.
Reference
Look-up material: vocabulary, coverage, and sources.
- Failure taxonomy The failure classes an agentic system must route deliberately.
- Concept coverage matrix Every concept in the course, mapped to the module that teaches it.
- Primary source map Every implementation-sensitive claim, mapped to its official source.
- Glossary Production-oriented definitions for the whole vocabulary.
Assessment
How work in this course is judged, including by you.
Capstone
The specification and brief for the system you build.