Gateaux - made to share

Gateaux is a data science platform for analytical work that needs to be shared, repeated and kept up to date.

Interested in using Gateaux for your team?Get in touch
Already have an account with us?To log in

Gateaux brings together code, data, software environments and workflows, so teams can collaborate on analysis and reproduce it without rebuilding the work each time.

Built by data scientists for data scientists, Gateaux supports the way analytical teams actually work, while making it easier for organisations to manage, run and trust the work they depend on.

Diagram of a reproducible analytical workflow, using a cake-making metaphor
Why Gateaux

Designed for modern data science

Gateaux supports analytical work from development through to repeatable production workflows.

Work in the tools and languages that suit the analysis, use the compute needed for the job, and bring different specialists into the same workflow. Gateaux keeps the workflow and its history together as the work develops.

  • Reproducible

    Run the same workflow again using the same code, data and software environment.

  • Auditable

    Trace an output back through the workflow that produced it.

  • Transparent

    See the code, data, assumptions and steps behind the analysis.

  • Automated

    Run recurring work on demand, on a schedule or when a trigger occurs.

  • Collaborative

    Let specialists contribute to one workflow without rebuilding each other's work.

  • Flexible

    Use the analytical tools that suit the work, while keeping one shared way to run and record it.

How it works

A dynamic hub for all your input and output

Gateaux connects the parts of an analytical workflow, from approved data and versioned code through to the outputs people use. The whole workflow can then be rerun, reviewed and shared.

A diagram of the types of input, and the types of output expected from gateaux

Platform architecture

Gateaux runs reports as batch jobs in cloud compute. Reports are linked to a code repository (hosted on either GitHub or GitLab) and a branch, and are directories with a configuration file. This file specifies a docker image, compute requirements, and any upstream dependencies.

Report runs, or jobs, can be triggered manually or when code is pushed.

The jobs are queued and run in runners on cloud compute, with the report code being executed inside the specified docker image. The results are stored in object storage, and are available for download from the web interface.

Organisations may provision runners inside their own infrastructure so that they have full control over the compute and storage.

A diagram of gateaux's architecture, including services used
Who is it for

One workflow, different expertise

Gateaux supports the way multi-disciplinary analytical teams work.

  • Data engineers
    Connect and prepare data, and manage data pipelines.
  • Data scientists and modellers
    Develop, test and run analysis and models.
  • Researchers and subject experts
    Review methods and assumptions, interpret results and contribute domain knowledge.
  • Report and visualisation specialists
    Turn analysis into reports, visualisations and other usable outputs.
  • Managers and reviewers
    See what ran, when it ran and which outputs are current.
Features & benefits

Built into Gateaux

Gateaux is designed to make the most of new technology while using it responsibly. Teams have flexibility in how they work, while organisations retain control over their data, compute, costs and security.

  • Responsible AI-assisted analytics

    Use AI alongside analytical workflows while keeping the underlying work visible and traceable.

  • Carbon-conscious computing

    Workflows are designed to use compute, memory and storage efficiently. Gateaux development is undertaken using renewable energy.

  • Agnostic by design

    Use the analytical tools, languages and infrastructure that suit the work, rather than being tied to a single technology stack.

  • Controllable compute

    See and manage compute use, set limits to stay within budget, and choose the resources appropriate to each workflow.

  • Secure by design

    Gateaux is designed to support secure analytical environments, with access controls, protected credentials and clear separation between users, data and compute.

  • Data sovereignty

    Gateaux is New Zealand owned, with data hosting available in New Zealand or Australia. Organisations can also deploy Gateaux within their own infrastructure where greater control over data residency is required.

Pricing

Flexible deployment

Choose the way Gateaux works best for your team and infrastructure.

OptionWhat it isPricing
Research and small teams
Use Gateaux directly for individual research or smaller team workflows.
Compute only
Organisation
A dedicated organisation environment for teams working together in Gateaux.
Maintenance & support + onboarding + compute
Enterprise
Gateaux deployed within your organisation's own infrastructure.
Maintenance & support + onboarding
Managed service
Dragonfly runs and maintains an analytical or modelling service for you, with Gateaux as part of the underlying capability.
Managed service + compute

Stories about Gateaux

newsGateaux T-shirts coming soon!A picture of two gateaux t-shirt designs
newsThe Kākāpō Recovery Program is now using GateauxA picture of a kākāpō

Organisations who are using Gateaux

  • Fisheries New Zealand
  • Department of Conservation
  • SRPFMO
  • Kahawai Collective

Since using Gateaux my life has changed. Nullam id dolor id nibh ultricies vehicula ut id elit. Duis mollis, est non commodo luctus, nisi erat porttitor ligula, eget lacinia odio sem nec elit.

Jerome McMillanChief Patissier, MBIE