Our work
Rogue Science’s secret weapon is our in-house software, Everest.
Everest is meta-modelling suite, capable of tackling any modelling task. This is possible either through ‘engines’ or interpolation of a model into Everest’s schema.
Engines allow us to run complex, agent-based models at scale, more efficiently (and, therefore, cheaply) than other modelling software frameworks for specific types of problems.
Learn more about our existing Everest Engines and projects.
So, what makes modelling with Everest different?
Model anything¶
Many modelling software packages, whether open source or commercial are domain-specific. Everest, by design, can be applied to any problem, in any sector.
Scale by design¶
By throwing the rulebook for model-building out the window, Everest helps analysts build and run hundreds, thousands, or millions of related simulations efficiently and cost-effectively.
Portable, reproducible, transparent¶
Everest becomes more efficient as more models are run. Rather than running a single simulation at a fixed price, with another bolt-on cost to edit or run the model again, once an Everest model is built and run, every component of the model continues to ‘exist’ as a readdressable end-point in the Everest Uniplex file system - ‘running’ it again becomes trivial.
Progressive complexity¶
Everest’s modular design means sub-models can be integrated into a parent model. As a research question evolves, and the ‘landscape’ of the problem reveals itself, a relatively simple model can scale in complexity by ‘snapping’ in sub-models, all without breaking compatibility or requiring any retooling of input/output control flow.
Ask meta questions of a model¶
Modelling is often focused on providing a ‘solution’ or an ‘optimum outcome’. Everest inverts that, encouraging researchers to ask bigger, broader, and wider questions of their model. Rather than running a scenario to see the outcome, a researcher can ask “what range of scenarios would give me a certain outcome”.
Work with us¶
Can Rogue Science support your research goals? Get in touch