AI Agents in Engineering, NASA CFD & CAD Prediction Using AI

🔩 "Engineering is the art and science of nuts and bolts.” – Haresh Sippy

Jul 06, 2025

💻 My Talk for Siemens at RealizeLive in Amsterdam

Presenting at Siemens RealizeLive and talking about applications of AI Agents in Engineering was a lot of fun!

I presented the future of Agentic AI, using Synera as an example of one of the leaders in the space, and drew parallels to n8n that I am using on a daily (!) basis, highlighting the bottlenecks of RAG systems and the anterograde amnesia of LLMs.

🔴 Upcoming LIVE SESSION: How AI-Driven Simulation Transforms Engineering Workflows

Simulation is traditionally time-consuming, expensive, and limited by manual iteration. In this live deep dive, we’ll walk you through StrömungsRaum®, a next-gen simulation platform that radically accelerates engineering design with AI-powered workflows and real-time optimization.

From problem definition to final design export, you’ll see how StrömungsRaum® blends physics-based high-fidelity simulations with surrogate modeling and intuitive UI tools - delivering rapid iteration, smart variant comparisons, and clear decision support for engineers.

📝 CFD for Injection Molding

This white paper highlights current challenges in injection molding and introduces new simulation techniques with StrömungsRaum®, which enable more realistic and efficient processes. Application examples, technological advantages as well as company and market insights round off the overview.

Download for Free

💻 Open Source Multiphase Flow Modeling for Real-world Applications

Official Website

This open-source software has over three decades of development history and more than 7,000 registered users worldwide. MFiX has become the standard for comparing, implementing, and evaluating multiphase flow constitutive models and has been applied to an extremely diverse range of multiphase flows applications. The successes achieved in modeling complex multiphase flow systems have led to new and improved key attributes such as drag, polydispersity, attrition, and agglomeration models, among other significant advances.

💻 CFD Lectures for Incompressible Flow

The present lecture notes are written to emphasize the mathematics of the Navier–Stokes (N.–S.) equations of incompressible flow and the algorithms that have been developed over the past 30 years for solving them.

Full Lecture Notes

🧠 AI Helps Predict and Sketch Computer-Aided Design Models

Parametric computer-aided design (CAD) is the dominant paradigm in mechanical engineering for physical design. Distinguished by relational geometry, parametric CAD models begin as two-dimensional sketches consisting of geometric primitives (e.g., line segments, arcs) and explicit constraints between them (e.g., coincidence, perpendicularity) that form the basis for three-dimensional construction operations. Training machine learning models to reason about and synthesize parametric CAD designs has the potential to reduce design time and enable new design workflows.