Loading content
This course includes: 13 lessons · 4 supporting documents
Pre-course assessment
Check your starting knowledge before learning
What are AI Agents?
Lesson video
Choose a suitable process
Automation suits clear, recurring steps with accessible inputs and verifiable results. Map the current process first; separate rule-based work from judgment and assess time saved, maintenance cost, and the cost of errors.
Build the flow and handle exceptions
Define triggers, inputs, conditions, outputs, owners, and handling for missing data or outages. Consider duplicates, invalid formats, cancellations, and timeouts; specify what happens after partial completion.
Decide whether to use rules or AI
Use fixed rules for required-field checks or status notifications. AI may classify or draft, but is uncertain. Define when a person must review; never let a guess automatically decide rights, money, or outcomes for individuals.
Design the approval step
Show context, impact, sources, and ways to reject or request more information. Approvers need authority for the task; separate creator and approver for high-risk work. Define that the process pauses when there is no response, a timeout, or a rejection.
Test and monitor
Test normal and failure cases, including empty or invalid data, duplicates, insufficient permissions, and outages. Start small with a manual fallback, measure errors and downtime, and assign alert recipients and workflow owners.
Applied case study and evidence review
Prevent duplicate work when the same request is delivered again
Course study guide
A practical guide to the course lessons, with Thai and English editions.
Practical workbook
Course-based exercises for review and planning a small real-world trial.
Quick reference
A concise reference to course topics and review points for use while working.
Implementation planner
A template for connecting course topics to goals, owners, and evidence of results.
Post-course assessment
An 8-question quiz to review your understanding.
Loading comments…
