Register for AI / ML for Fluids: A Cohort
Second Cohort Starts · 10 Nov 2026
Limited seats only — secure yours to keep this event interactive
Delivered by leading experts in AI / ML for fluids
A live, expert-led programme — built for working engineers
Format
Real-time interactive sessions · limited seats
60+ hrs
Total Hours
Of live, hands-on instruction
3 months
Duration
Starts 10 Nov 2026
Certified
Outcome
In one of the most in-demand skills today
Applied
Approach
Theory + practical · industry use-cases
Prof. R. Vinuesa
Led by
From University of Michigan · + 3 expert instructors
Why this cohort, why now?:
The intersection of machine learning and computational fluid dynamics is moving fast — and it’s moving from research labs into industry. Top journals are publishing it, national labs are funding it, and engineering teams across aerospace, automotive, energy, and biomedical sectors are actively hiring for it. But the talent pool is still small. Which is exactly why now is the right time to step in.
This cohort is designed for anyone with a basic understanding of programming and CFD — students, researchers, and working professionals in the field. You don’t need prior experience in AI or machine learning. We start from scratch and take you all the way to advanced topics, including physics-informed neural networks, neural operators, and surrogate modelling for real engineering problems.
If you’re a student or researcher, this is one of the most promising niches you can specialise in today. The field is still wide open, the demand is growing, and being early matters — in PhD applications, in interviews, and in the job market that’s shaping up around it.
If you’re a working CFD engineer, the question isn’t whether ML will reshape your field — it already is. Surrogate models are cutting simulation time by orders of magnitude. Physics-informed networks are solving inverse problems. Foundation models for turbulence are emerging. The engineers who learn this now will be leading those projects in two years.
The programme covers the full path: theory you actually need, the programming and libraries used in the field, and hands-on practice on real-world problems — not academic toys. You’ll learn from a faculty led by Prof. Ricardo Vinuesa (University of Michigan), supported by industry experts who apply these methods in their day jobs.
The first cohort starts on 10 Nov 2026. Seats are limited.
Learn from leading academics and industry practitioners
A faculty led by world-class researchers in AI for fluids, joined by industry experts who apply these methods to real engineering problems every day.
Prof. Ricardo Vinuesa
Lead InstructorFrom University of Michigan
A leading researcher at the intersection of fluid dynamics, AI, and sustainability. Associate Chair for Research and Associate Professor of Aerospace Engineering at the University of Michigan, with over 13,000 citations and ~200 publications applying deep learning, reinforcement learning, and explainable AI to turbulence and flow control.
Prof. Vinuesa studied Mechanical Engineering at the Polytechnic University of Valencia and earned his MSc and PhD at the Illinois Institute of Technology, Chicago. Before Michigan, he was Associate Professor at KTH Royal Institute of Technology in Stockholm, where he served as Vice Director of the KTH Digitalization Platform and Lead Faculty at the KTH Climate Action Centre.
His landmark 2020 paper in Nature Communications on the role of AI in the UN Sustainable Development Goals influenced the EU AI Act and has been highlighted by UNESCO as a key contribution. His honors include an ERC Consolidator Grant, the Göran Gustafsson Award, the TSFP Kasagi Award, the IIT Outstanding Young Alumnus Award, and membership of the Young Academy of Science of Spain.
Prof. Sergio Hoyas Calvo
From Universitat Politècnica de València
Full Professor of Aerospace Engineering at UPV (Spain) and an internationally recognised expert in turbulence, CFD, and high-performance computing. Holds a degree and PhD in Mathematical Sciences from the Complutense University of Madrid, with a long-standing focus on numerical methods for complex fluid mechanics problems.
Prof. Hoyas began his academic career across several Spanish universities and was appointed Full Professor at UPV in 2019. His research applies supercomputing and direct numerical simulation (DNS) to wall-bounded turbulence, with widely cited work on turbulent boundary layers and turbulent Couette flow. He is also a Lecturer in Mathematics for the international programme of Florida State University.
More recently, he has integrated explainable AI into the study of turbulent dynamics, including collaborations identifying regions of importance in wall-bounded turbulence through deep learning. He is a frequent co-author of Prof. Vinuesa, and brings to the cohort decades of expertise translating advanced mathematics and HPC into practical CFD methodology.
Dr. Azeddine Rachih
From Eramet · Paris
Senior Modeling Engineer at Eramet, applying AI and physical modeling to industrial metallurgy and engineering systems. PhD in engineering with prior research experience at CEA, and a strong foundation in fluid mechanics, heat transfer, and numerical methods from École Centrale Paris and CentraleSupélec.
In this cohort, Azeddine leads the industrial problem-solving track — translating CFD challenges from real engineering settings into ML and AI workflows that work in practice. His day-to-day work spans physical modeling, digital metallurgy, and applied machine learning for production environments, giving cohort members a direct view of how these methods are deployed beyond academic papers.
Nishant Soni
From AMD
Senior Software System Designer at AMD, working on accelerating scientific applications with AI and HPC. Previously a CFD Applications Engineer at COMSOL, with a postgraduate certification in AI / ML from Caltech and an M.S. in Computational Mathematics from JNCASR Bangalore.
In this cohort, Nishant supports participants on the programming and tooling side — covering Python, key libraries such as TensorFlow, PyTorch, and pandas, and core concepts in HPC and AI. His combined background in CFD, parallel programming (MPI / OpenMP), and applied machine learning helps engineers move from fluid mechanics fundamentals to working confidently with the modern AI / ML toolchain.
Schedule & Times
Most live sessions take place on Saturdays and Sundays at 8:00 AM ET / 1:00 PM UK / 6:30 PM IST — chosen to work across the US, Europe, and India. Each session runs for 2 hours with interactive Q&A.
Some weekday sessions will also be scheduled, and exact timings may shift slightly depending on instructor availability. The full schedule is shared at orientation, before the cohort starts on 10 Nov 2026.
All sessions are recorded and shared with cohort members, so you can revisit any topic and never miss a class.
Course content in detail
Curriculum / What You’ll Learn
The cohort takes you from the foundations of programming and AI/ML through to physics-informed deep learning, neural operators, and real industrial CFD applications — structured across four phases plus a closing project phase.
INTRODUCTION SESSION
- Introduction to AI and ML and overview of the cohort
- Logistics, plans, and schedule
PHASE 1: PROGRAMMING SKILLS, TOOLS, AND FOUNDATIONS
- Python essentials: setup, array manipulation, publication-quality flow plots — theory + practicals
- Core scientific libraries: NumPy and Pandas — theory + practicals
- Google Colab environment setup for all sessions (free GPU access, with upgrade options)
- Introduction to TensorFlow and PyTorch — theory + practicals
- Foundations of neural networks: artificial neurons, activation functions, loss functions — theory + practicals
- Loading real CFD output; computing flow statistics; visualising velocity and vorticity fields — theory + practicals
- Feature engineering: handling real CFD file formats, cleaning large datasets, computing turbulence statistics — theory + practicals
- HPC, GPU, and computing concepts relevant to AI/ML workloads
PHASE 2: DEEP LEARNING & NEURAL NETWORKS
- Neural networks from perceptron to Keras: MLP fundamentals — theory + practicals
- CNNs for spatial flow field analysis — theory + practicals
- RNNs and LSTMs for temporal flow prediction — theory + practicals
PHASE 3: ML FOR CFD, PINNs & NEURAL OPERATORS
- Reduced-Order Models: POD, DMD, and autoencoders — theory + practicals
- Industrial CFD + ML workflows: real-world case studies from industry — theory + practicals
- Transformers and attention mechanisms for flow fields — theory + practicals
- Physics-Informed Neural Networks (PINNs): embedding Navier-Stokes into the neural network loss; solving forward and inverse fluid problems — theory + practicals
- Neural Operators: FNO and DeepONet — theory + practicals
- Optional: Advanced topics on the latest developments in AI/ML for fluids
CLOSING PHASE: PROJECTS & DATASETS
- Open-source datasets from the literature with implementation guidance
- Additional industry-relevant datasets
- Concluding session and Q&A
Frequently asked question
- Who can participate in this program?
This cohort is designed for engineering students, researchers, and working CFD professionals who want to learn AI and ML applied to fluid mechanics. We start from the basics of programming and AI/ML, so you don’t need prior experience in machine learning. It’s particularly suited for CFD engineers looking to upskill into one of the most in-demand niches in engineering today. - What is the mix of students?
The cohort brings together a diverse, international group — bachelor’s, master’s, and PhD students alongside engineers and researchers from industry. You’ll learn alongside peers from across the world working in similar domains. - Are there any prerequisites?
A good understanding of fluid dynamics fundamentals, basic CFD knowledge (familiarity with generating data using CFD tools), and basic programming awareness will help you get the most out of the cohort. That said, the programme is designed to take you from fundamentals to advanced topics, so deep prior experience in AI/ML is not required. - Do I need programming knowledge to join?
Some basic programming familiarity is helpful, but no advanced experience is required. The cohort begins with Python foundations and key scientific libraries (NumPy, Pandas, TensorFlow, PyTorch), so you can start at the basic level and build up. - Do I get a certificate?
Yes. Upon completion — submitting one project report and reports for any three practicals discussed during the cohort — a verified certificate is issued by Flowthermolab. - Can I pay in installments?
Yes. Flexible payment options are available at checkout — Razorpay with EMI (up to 12 months) for India, and Klarna via Stripe for international payments - Are there discounts for group enrollments?
Yes. We offer special pricing for groups of 3 or more from a company, lab, or research group. Email support@flowthermolab.com with the number of participants and your company/institute details, and we’ll send you a quotation and invoice. - Do I need a powerful workstation or computer for this course?
No. A standard laptop is sufficient. All practical sessions use Google Colab (free tier works fine), with the option to upgrade for higher GPU usage if needed. Heavy compute is handled in the cloud. - Do I need licensed software?
No. The cohort uses open-source tools and freely available platforms — primarily Python with libraries like TensorFlow, PyTorch, and pandas, all running on Google Colab. Datasets and codes are shared with participants. - What if I get stuck on a topic?
All live sessions include interactive Q&A, and recordings are shared so you can revisit any topic. Industry mentors and instructors are available through the cohort to answer questions and help you work through difficulties. - Will I get access to the session recordings?
Yes — every live session is recorded and shared with cohort members. So if you miss a session due to time-zone or schedule conflicts, you can catch up on your own time. - When does the cohort start?
The Second cohort starts on 10 Nov 2026 and runs for approximately 3 months, with around 60+ hours of live instruction. Sessions are mostly on Saturdays and Sundays at 8:00 AM ET / 1:00 PM UK / 6:30 PM IST
Who is this course for?
- Bachelors, master’s, and PhD students in Mechanical, Aerospace, or related engineering disciplines, looking to specialise in one of the most in-demand niches in modern engineering
- Research scientists working in fluid mechanics who want to integrate AI and ML into their research workflows
- CFD engineers in industry ready to upskill and lead the next wave of ML-driven simulation projects in their organisations
- Computational engineers and applied mathematicians with a fluids background looking to add machine learning to their toolkit
Join the Second BATCH cohort of AI / ML for Fluids
One-time fee of $500 USD covers the full programme — 60+ hours of live instruction, all session recordings, hands-on practicals, shared codes and notebooks, and certificate of completion.
Enroll now to secure your seat and get free access to two recorded Advanced Python courses as a bonus, so you can prepare before the cohort starts on 10 Nov 2026. All cohort codes and notebooks are also shared with participants.
Only a limited number of seats remain. Confirm your participation early and be part of a cohort with engineers and researchers from across the world.
Instalment options available at checkout: select Razorpay and choose EMI (up to 12 months) for India payments, or Stripe and choose Klarna for international payments.
Questions or need to discuss? Email support@flowthermolab.com
Enrolling a group? Special discounts available.
Bringing colleagues from your company, lab, or research group? Special pricing for groups of 3 or more. Write to support@flowthermolab.com with the number of participants and your company or institute details (name, address, billing contact), and we'll send you a quotation and invoice.