Course C2 — Modern Engineering Workflows: Python, AI Agents and Cloud

2–3 days (Day 3 optional GCP deep-dive) | For engineers running FEM analyses | English

Overview

This course teaches engineers how to automate, scale, and accelerate their numerical modelling workflows using Python, AI agents powered by Claude, and Google Cloud Platform. It is not a generic programming or AI course — it is built specifically around the engineering workflows that participants already run (OpenSees analyses, parametric studies, result post-processing, report generation) and shows how to replace manual, fragile, repetitive processes with robust automated pipelines. The AI agent component demonstrates how Claude can be used to manage simulation runs, interpret convergence failures, generate code, and produce structured reports from raw results.

Prerequisites

Basic Python, familiarity with at least one FEM tool (OpenSees, ABAQUS, PLAXIS, etc.).

Software

Software: Python (NumPy, Pandas, Matplotlib, Jinja2), Claude API, Google Cloud Platform (Compute Engine, Cloud Storage), GitHub

Learning Objectives

Build parametric OpenSees model generators in Python — no more manual TCL editing
Automate result parsing, damage measure extraction, and data aggregation across large result sets
Generate professional analysis reports automatically using Python templating
Design and implement AI agent workflows using the Claude API for engineering tasks
Use AI agents to manage simulation queues, interpret log files, and debug convergence failures
Set up and run FEM jobs on Google Cloud Platform VMs with cost monitoring
Implement cloud storage pipelines for large parametric study result management

Daily Schedule

Day	Topics
Day 1	Python for engineering automation: parametric model generation, result parsing pipelines, data aggregation with Pandas, automated plotting, report generation with Jinja2 templates. Best practices: version control with GitHub, reproducible environments.
Day 2	AI agents in engineering: Claude API fundamentals, designing agents for simulation management, prompt engineering for technical tasks, building an agent that reads OpenSees logs and diagnoses convergence failures, automated code generation and review workflows.
Day 3 (opt.)	Google Cloud Platform for engineers: VM setup and configuration, SSH and file transfer, running OpenSees jobs at scale, Cloud Storage for result management, cost monitoring and budgeting, combining GCP with automated Python pipelines.