top of page

Deliver your own Bayesian network applications in the cloud

Build more than just models. enables data scientists, analysts and AI engineers to easily create and deploy models as cloud-based applications for up to thousands of users.


Using the latest developments from the fields of artificial intelligence, causal modelling and Bayesian probabilistic reasoning, supports AI-driven applications across the enterprise. is simple to deploy and scale.


"Let's be honest, most risk assessment methodologies are guesses, and not very good ones at that. People collect statistics about what they can see and then assume it tells them something about what they can't." 

Neil Cantle, Principal, Consulting Actuary, Milliman LLP is engineered to make it easy to deploy AI applications in the cloud

The modeller is a 'no-code' design and execution environment for creating Bayesian networks and causal models which runs on Windows, Linux and Macintosh operating systems. modeller delivers state of the art algorithms for Bayesian and causal model computation coupled with an easy to use interface.


Models developed in modeller can then be deployed to the cloud environment as web applications or as computational APIs.


The cloud service includes a computational cloud API that executes models developed using modeller, and a design environment to create and deploy Web Apps to your end users.

The cloud service is a hosted service that uses the latest technologies including Kubernetes and Kafka.


We can also support your private cloud where required to deploy your application.

Application showcase

Alzheimer's diagnosis

download.jpeg can be used for health applications, for diagnosis, risk assessment and prognosis.


To try the Alzheimer's disease diagnostic WebApp click this button and it will open a new tab in your browser to launch the application.

Cyber-security risk assessment

Image by Joshua Woroniecki can be used for cyber security risk evaluation.


To try the Cyber-security risk assessment WebApp click this button and it will open a new tab in your browser to launch the application.


R.jpg's modeller API is available for Java, Python and 'R'. Our modeller API can be run locally on a single machine or can be connected to the cloud API for remote computation. The modeller API provides the same core functionality as the modeller without the graphical user interface.


AI models, data input and data output are provided, with easy-to-use JSON formats as well as CSV files.


The modeller API is designed for use by data scientists and programmers to makes it easier to develop more complex models and to automate data analysis and model execution.

Licensing options

academic inverted.JPG

Academic research and class licenses are available at a substantial discount. 


Book an introduction call or demo with one of our team:

Technology's Bayesian technology is based on innovative research in computer science, AI, causal reasoning,Bayesian probability, and data analysis. It has been engineered to help organisations make smarter decisions. helps model problems when you have data but also improves decision making where data is sparse, where direct measurement is not possible, and in the face of new, often novel, circumstances. Our technology can be used in multiple industries and meets the demands of different types of users and organisations.

'No-code' solution to build and compute AI models:

  • Graphical UI to build graph of model structure.

  • Computer Bayesian network model for inference: prediction, diagnosis and causal explanation.

  • Create multiple node types: Boolean, Continuous, Labelled, Ranked, Discrete Real.

  • Enter and withdraw soft and hard evidence.

  • Support for hybrid models containing discrete and continuous variables.

  • Attractive and flexible charts for displaying multiple scenarios.

  • Algorithms include Junction Tree and Dynamic Discretization and Expectation Maximisation.

  • Compound sum analysis.

  • Value of information analysis

  • Sensitivity analysis.

  • Multivariate analysis.

  • Hybrid Influence Diagrams.

  • EM-based table learning from data and expertise.

  • Statistical distributions: Normal, Triangular, Beta, Gamma, etc.

  • NoisyOR, NoisyAND, mFROMn.

  • Arithmetical operators +, -, *, /, ^.

  • Comparative expressions, IF(x < 10, "True",....).

  • log( ), ln( ), sqrt( ), e, pi, mod( ), min( ), max( ).

  • Comparative operators: ==, !=, <, >, >=/ <=, && (AND),  || (OR), XOR.

  • Ranked node functions.

Cloud portal:

  • WebApp designer to create web applications from models.

  • Cloud App manager to deploy webApps to internet domains.

  • Access control rules for end users.

  • Account management.

  • API service endpoints for model computation using case data.

  • Server to sever automated pipelines

  • Support for tailored applications.

WebApp UI designer

  • Load models JSON files.

  • Create WebApp application names and descriptions.

  • Select inputs and output variables for end user data input and for reporting & charts.

  • Organise inputs into groups for easy data entry.

  • Use charts for reporting: area, histograms, scatter and line plots.

  • Export results as pdf files.

  • Save and load case data files.

Cloud API

  • Calculate model endpoint.

  • Calculate specialized analyser endpoints.

  • Support for synchronous operations.

  • Asynchronous operations with callback.

  • Asynchronous operation with polling.

  • Authentication via access tokens.

  • Account management.

Special purpose analyzers:

Statistical distributions and functions:

About us

Our mission is to help you develop and deliver at scale your Bayesian network applications in the cloud.


Agena was founded by Professor Norman Fenton and Professor Martin Neil, who have published hundreds of papers and books on Bayesian Networks for AI and probabilistic reasoning.


Our software is used across a number of industry sectors, including defence, energy, financial services, health, telecommunications and transport, supporting a range of decision-making applications, including AI in healthcare, catastrophe modelling, cost benefit analysis, cyber security risk, infrastructure resilience and investment decision making.

bottom of page