AI-Augmented
CFD Workshop
Work smarter with AI.
Stay in control of the physics.
Turn AI into practical support for your CFD workflow — from case setup and troubleshooting to post-processing and technical review. Connect your tools, your knowledge and your engineering decisions.
Explore the workshop programmeFrom one-off prompts
to a connected workflow.
Three connected modules build on one another: give your assistant CFD context, ground its answers in technical sources, then coordinate tasks with engineering review. Guided examples connect each step to your simulation workflow.
- 01 / ASSISTYour CFD copilot
Give AI the context to help.
- 02 / GROUNDYour knowledge library
Connect answers to technical sources.
- 03 / COORDINATEYour AI agent team
Define roles, hand-offs and review.
Build your personal
CFD copilot.
Give an AI coding assistant the project context, instructions and boundaries it needs to support useful engineering work.
Work through case setup, scripting, diagnostics and post-processing. Learn to inspect proposed changes, test generated code and keep the original case traceable.
- Describe the flow problem, solver version, constraints and expected outputs.
- Generate and modify OpenFOAM case files from a known starting point.
- Support batch tasks, log checks and Python post-processing.
- Review changes and test outputs before accepting them.
The building block: reusable project instructions and an AI-assisted CFD task.
Build your CFD
knowledge library.
Put relevant manuals, papers, validation studies and project notes within reach of your assistant.
Explore Retrieval-Augmented Generation (RAG): retrieving relevant material before drafting an answer. Learn to check citations, separate evidence from inference and recognise when the available sources are not enough.
- Organise a small, relevant set of technical references.
- Ask source-grounded questions and inspect the retrieved passages.
- Identify version mismatches, missing context and unsupported claims.
- Explore local AI deployment and its privacy considerations.
The building block: a starter knowledge-library structure and a source-checking method.
Build your AI
agent team.
Break a larger workflow into clear responsibilities: planning, setup, knowledge retrieval, diagnostics, analysis and review.
Learn how specialist agents can exchange structured outputs and flag issues for the engineer. Start with a bounded workflow, explicit acceptance checks and approval before consequential changes.
- Define each agent’s task, inputs, permissions and expected output.
- Connect project files and technical references to the right role.
- Use a separate review role to challenge assumptions and changes.
- Bring results into a report with traceable evidence and open questions.
The building block: a small agent-team workflow with engineering review gates.
Tools we use
The demonstration environment connects case files, coding assistance and technical knowledge across the three modules.
- OpenFOAMCase setup & simulation workflow
- Python + VS CodeScripting, analysis & automation
- OpenAI CodexAI-assisted coding & project work
- OllamaOptional local-model demonstration
These tools form the demonstration environment; the workflow principles are transferable to other engineering software stacks.
Three building blocks.
Your next practical step.
Each module contributes a foundation you can adapt to your own CFD work: project instructions for your copilot, a source-checking approach for your knowledge library, and a reviewed workflow for your agents.
These are starting points developed through guided examples — not a production-ready autonomous CFD system. The individual follow-up helps you decide how to apply and extend them.
Then, one hour focused on your implementation.
Work one-to-one with an expert within 30 days of the workshop. Discuss your workflow, troubleshoot implementation challenges and identify practical next steps.
Your AI-supported
CFD workflow
- 01A copilot with context
Project instructions, task boundaries and a repeatable way to work.
- 02Knowledge linked to sources
A structure for references, retrieval and evidence checks.
- 03Agents with clear responsibilities
Defined roles, hand-offs and human approval points.
- +Your individual follow-up
Turn the three building blocks into next steps for your CFD work and constraints.
Learning outcomes · developed through guided examples and discussion
Bring your CFD experience.
Build your AI capability.
The workshop builds on your existing CFD judgement. It is for engineers and researchers ready to apply these building blocks to a simulation workflow they already understand.
A good fit if you want to…
- Spend less time repeating setup, scripting and post-processing tasks.
- Make technical knowledge easier to find and check.
- Move from isolated AI conversations to a structured engineering workflow.
- Understand where automation helps — and where review is essential.
What to bring to the day
Bring basic CFD knowledge, familiarity with simulation files and a workflow task you would like to improve. Some experience with OpenFOAM, Python or the command line is useful; prior experience building AI agents is not required.
This is not an introduction to fluid mechanics or CFD. If you are unsure about the fit, tell us about your background before booking.
A few practical details.
What does the workshop fee include?
The fee is A$1,699 per participant. It includes the one-day live online workshop, with a maximum of four participants, and a one-hour individual technical follow-up to use within 30 days of the workshop.
When does the workshop run?
We run this live online workshop every month. Specific dates are arranged according to enrolment and participant availability. Please contact us before purchasing to confirm the next suitable workshop date and your place. After purchase, email your order number to solutions@42cfdlab.com so we can coordinate your attendance.
Is this a machine-learning or AI solver course?
The focus is AI around CFD: assisting with case preparation, scripting, knowledge retrieval, monitoring, post-processing and reporting. Training surrogate models, neural operators or physics-informed neural networks is not the focus of this workshop.
Can we discuss my own project?
Your workflow questions are welcome, especially in the individual follow-up. Use non-confidential examples in the group session. Before sharing project files, confirm that you have permission to use them and that the chosen tools meet your data-handling requirements.
How does the individual follow-up work?
After the workshop, contact us to arrange your one-hour online session within 30 days. Bring your implementation questions, challenges and priorities so we can work through practical next steps together. Meeting times are agreed with you, subject to availability.