Generating a Coffee Cup Thermal Model from a Schematic with Dyad's Agentic AI

Description

Watch how Dyad's AI agent builds a complete thermal model from just an image! In this video, we demonstrate Dyad's agentic workflow by recreating a coffee cup cooling model originally developed by Dr. Clément Coïc—starting only from a schematic image and simulation plots. The AI agent: - Analyzes the thermal model schematic and identifies all components - Generates parameter calculation scripts with physical correlations - Implements the complete model in Dyad with documentation - Calibrates the model using Dyad Model Optimizer to match reference results - Creates custom visualizations and animations This video showcases how Dyad transforms model development workflows, from understanding physical systems through images to generating production-ready code. Whether you're working with hand-drawn schematics, PowerPoint diagrams, or standard system drawings, Dyad's AI agent can help translate visual representations into validated physical models. **Key Highlights:** - Image-to-model generation with automatic component identification - Automated parameter calculations with physical insights - Model optimization and calibration workflows - Custom visualization and animation generation - Complete Julia code generation with documentation **Previous Video:** Friction Brake Model Development with Dyad - https://www.youtube.com/watch?v=hIkbUBqi6sI **Reference:** Dr. Clément Coïc's Coffee Cup Thermal Modeling Series - https://www.linkedin.com/pulse/005-good-models-one-time-thing-dr-cl%C3%A9ment-co%C3%AFc-70kcf/ **Interested in Dyad?** Contact us at sales@juliahub.com --- ## Timeline [00:00](https://www.youtube.com/watch?v=mHz4r-CQYIU&t=0s) - Introduction to agentic workflows series [00:05](https://www.youtube.com/watch?v=mHz4r-CQYIU&t=5s) - Overview of Dr. Clément Coïc's coffee cup thermal model [00:38](https://www.youtube.com/watch?v=mHz4r-CQYIU&t=38s) - Setting up the AI agent with model schematic [00:56](https://www.youtube.com/watch?v=mHz4r-CQYIU&t=56s) - Agent analyzes the thermal model image [01:08](https://www.youtube.com/watch?v=mHz4r-CQYIU&t=68s) - Understanding component selection and model structure [01:34](https://www.youtube.com/watch?v=mHz4r-CQYIU&t=94s) - Generating parameter calculation scripts [02:19](https://www.youtube.com/watch?v=mHz4r-CQYIU&t=139s) - Adjusting parameters for double espresso [02:53](https://www.youtube.com/watch?v=mHz4r-CQYIU&t=173s) - Creating parameter calculation report [03:06](https://www.youtube.com/watch?v=mHz4r-CQYIU&t=186s) - Model implementation and initial simulation [03:30](https://www.youtube.com/watch?v=mHz4r-CQYIU&t=210s) - Reviewing generated Dyad code [03:48](https://www.youtube.com/watch?v=mHz4r-CQYIU&t=228s) - Adjusting heat transfer parameters [04:04](https://www.youtube.com/watch?v=mHz4r-CQYIU&t=244s) - Modifying mug heat capacity [04:25](https://www.youtube.com/watch?v=mHz4r-CQYIU&t=265s) - Rendering graphical model representation [04:36](https://www.youtube.com/watch?v=mHz4r-CQYIU&t=276s) - Adding hand-holding effect to model [04:58](https://www.youtube.com/watch?v=mHz4r-CQYIU&t=298s) - Simulating model with hands [05:06](https://www.youtube.com/watch?v=mHz4r-CQYIU&t=306s) - Comparing results with reference data [05:31](https://www.youtube.com/watch?v=mHz4r-CQYIU&t=331s) - Model calibration using Dyad Model Optimizer [06:21](https://www.youtube.com/watch?v=mHz4r-CQYIU&t=381s) - Creating custom animations [06:42](https://www.youtube.com/watch?v=mHz4r-CQYIU&t=402s) - Rendering final model with hands [07:00](https://www.youtube.com/watch?v=mHz4r-CQYIU&t=420s) - Testing with abstract schematic [07:43](https://www.youtube.com/watch?v=mHz4r-CQYIU&t=463s) - Conclusion and future capabilities #Dyad #JuliaLang #PhysicalModeling #AIAgent #ThermalModeling #ScientificComputing #ModelingSimulation #JuliaHub #SciML #EngineeringSimulation #AcausalModeling #MachineLearning #ModelOptimization #ComputationalEngineering #DigitalTwin