Building No Code AI Agents for Engineering Practice
- Course Number: C1075
- Credits: 3 hours
- Instructor: Ellen C. Huang, PE
- Price: $30
Course Outline
This course teaches a reusable method for building, testing, and maintaining engineering AI agents without writing code. The hands-on lab uses Microsoft Copilot Studio to build an agent that screens construction requests for information (RFIs) and drafts routing notes for human review. The design method applies across agent-building platforms. You will define a task, supply approved references, limit what the agent can do, and test its decisions. The screenshots and setup steps are specific to Copilot Studio. The same method can support commissioning, submittal review, field reporting, and structural coordination.
A multiple-choice quiz reinforces the course material.
Learning Objectives
At the conclusion of this course, the student will:
- Distinguish a one-time AI prompt, a fixed workflow, and an agent that can select among allowed steps..
- Define an engineering agent’s scope, input requirements, source material, outputs, and escalation rules.
- Configure a no-code agent with plain-language instructions and a controlled knowledge source.
- Configure and test a text-only tool, then verify when the agent uses or bypasses it.
- Evaluate draft outputs for source support, accuracy, missing information, and unsafe behavior.
- Plan a limited pilot with human review, version control, and a documented stop procedure.
Course Content
You must login and purchase this course or a package to access this course content.