AI for Civil Engineering — Level 3: Expert
This level is for engineers, site supervisors, quantity surveyors and final-year students who already use a second brain and document templates (Level 2). You learn to map a real civil engineering workflow, such as the daily site report, the preparation of a bill of quantities or the tracking of watch points on sandy soil in a coastal area. For each step, you decide whether it stays human, whether AI assists it or whether it is automated. You then automate the repetitive steps with the no-code features of common tools. You build specialised agents, each limited to one precise task. You measure the time saved before and after, and you stay in control: a checking step, human approval and a way to stop. The agent prepares, the engineer decides: no decision with legal, financial or safety consequences is ever automated. No code is required. The course lasts about six weeks.
- 01
Map your workflows and spot what can be automated
You break down a real civil engineering task step by step: who does it, what goes in, what comes out and how long it takes. You measure the starting time, then sort each step into one of three categories: human, AI-assisted or automated. Finally, you choose the first workflow to work on.
4 lessons40 min - 02
Automate repetitive tasks on site and in the design office
You learn the structure of any no-code automation: a trigger, steps, an output. You apply it to the site report and to the formatting of quantity take-offs and the BOQ, using your Level 2 templates. Tools are described in a generic way: an automation tool, a shared folder, an assistant. You also check the data that moves through the workflow.
4 lessons42 min - 03
Specialised agents for precise tasks: instructions, knowledge, limits
Using an assistant-building tool, you build agents limited to a single task. Each one is written out in full on a sheet: its instructions, the knowledge it may use, what it must refuse and the person who checks its results. The examples cover site reports, quantity take-offs and watch points in coastal areas.
4 lessons42 min - 04
Measure the time saved and keep control
Every automation and every agent comes with four elements: a before/after measurement that you record yourself, a checking step, a human approval point and a way to stop it. You learn to count the time saved honestly, including the checking time.
4 lessons40 min - 05
Final project: an automated workflow with a specialised agent, measured and under control
You carry out, from start to finish, the automation of a workflow from your own job, together with a specialised agent: workflow map, starting time, automation, agent sheet, before/after measurement, checking, human approval and a way to stop. You bring everything together in a one-page file to present to your manager.
4 lessons42 min