The EASC is dedicated to advancing aerospace research
though smarter high performance computing.
The EASC provides the computing building blocks so that research teams can explore ideas faster
and more cost efficient.
Platform Features
Collaborative AI modelling
Work together in a secure browser-based IDE to build, test, and improve models. Teams can use shared code, notebooks, datasets, and project environments without needing to manage separate local setups
Dataset preparation
Import existing datasets and use the AI assistant to help clean, structure, validate, and prepare them for modelling. This can include identifying missing values, formatting issues, inconsistencies, and preparing the data for training or evaluation.
Model import and testing
Bring in existing models, open-source models, or internally developed models and run them inside a controlled workspace. Teams can test how models perform on their own datasets and compare outputs in a reproducible environment.
Experiment execution
Run experiments on dedicated GPU infrastructure without needing to configure the compute environment manually. Users can launch model runs, track progress, review logs, and store outputs in the same workspace.
Results understanding
Use the AI assistant to help interpret outputs, summarize findings, compare model performance, and turn raw results into clear explanations that can be shared with colleagues or used in reports.
Possible Use Cases
Surrogate modelling
Train AI models that approximate complex aerospace simulations or experimental systems. Surrogate models can help teams explore design options faster, reduce dependence on expensive simulation runs, and support early-stage decision-making.
CFD/FEA result learning
Turn high-cost CFD and FEA outputs into reusable predictive models. By learning from previous simulations, teams can estimate key performance indicators more quickly and reserve full simulations for validation and critical cases.
Design-space exploration
Automatically run, compare, and analyze many design variants across a defined parameter space. This helps researchers identify promising configurations, understand trade-offs, and accelerate the path toward better-performing designs
Digital twin experimentation
Update and refine models as new simulation, test, or operational data becomes available. This enables teams to continuously improve their understanding of system behavior and experiment with future scenarios in a controlled environment.
Physics-informed machine learning
Combine aerospace domain knowledge with modern AI techniques to create models that are more interpretable, constrained, and scientifically meaningful. This approach can help improve generalization and reduce reliance on purely data-driven methods.
Don’t know where to start?
Our learning center guides you through every step of setting up your workspace.