Meet engicloud.ai
engicloud.ai is a Python-based modeling and simulation platform for engineers and scientists, and the newest product in the DCS Computing family. It runs in a browser tab, with nothing to install locally and no license to manage.
The interface is a graphical canvas that works like a block diagram. Connect models, run them, and iterate as you go.
Make live API calls to databases, spreadsheets, LLM and ML tools, and other simulation software, and use the results inside your own models. Anyone you share a project with can run it immediately.
Why we built it
Engineering knowledge is scattered by default: textbooks, papers, a colleague's file, institutional memory that walks out the door when a person does. Very little of that is executable, and almost none is auditable in the form it's kept.
LLMs have sharpened that problem. As they write code faster than humans ever could, the question shifts from generation to validation. A model that returns a plausible number and can't be checked is worth very little to an engineer signing off on a design.
So our answer is to decompose complexity into verified building blocks and validate each component with unit tests. Most models trace back to a cited equation, and the source stays explicit, so the result is checkable.
Open a model, read the Python code, check the assumptions and the units, see where a number came from. Nothing sits behind a solver you can't inspect.
What's underneath
The building block of engicloud.ai is a Calculator: a Python function with typed inputs and outputs, units, documentation, and version history.
It can hold a physics-based model, an ML model, or an API call to a database, a spreadsheet, an LLM, or another simulation tool. Experts write the code; anyone on the team runs it from the browser. Connect Calculators on the canvas, and you have a project.
You don't start from an empty canvas either. More than 20,000 ready-to-use equations and models are already on the platform. Find yours with Semantic Search: describe the problem in plain language and it matches on meaning, not strings. Import Python you already have and it becomes a Calculator.
Find models in your industry
The library covers 20 engineering fields. Three examples to explore today:
- Attrition Model. Simulates how needle-shaped API crystals break down during agitated drying, tracking length and width separately through a 2D population balance. Developed through the Enabling Technologies Consortium with Bristol Myers Squibb, AstraZeneca, GSK, Genentech, Merck, and Pfizer.
- Drug Dissolution Model. A spherical drug particle dissolving, solved as a 3D transient diffusion problem rather than fitted to a curve.
- Heart Attack Predictor. A statistical model over medical parameters, on the same canvas, with a data-driven block instead of a mechanistic one.
Open one, change an input, and everything downstream moves.
Have a Python modeling task or idea?
We're glad to discuss and help you build it.
The Author:
Dr. Christoph Kloss
Co-founder and director of DCS Computing.