Enabling cross boundary data analytics.

Hazy focuses purely on synthetic data generation for enterprise data analytics. We specialise in preserving signal and relationships in customer information and transactional financial services data.

For enterprise financial services.

Enterprise-class capability

A genuinely enterprise class capability and track record of successfully enabling real world enterprise data analytics use cases.

Time series data support

Advanced generative models that can actually preserve sequential and causal relationships in transactional and time series data.

Formal privacy guarantees

Formal differential privacy guarantees that can be configured on a use-case by use-case basis to optimise the privacy vs utility trade-off.

A new paradigm in information transfer

Hazy uses advanced generative models to distill the signal in your data before condensing it back into safe synthetic data.

Synthetic data generation and transfer illustration

1. Train

Hazy uses generative models to understand and extract the signal in your data.

2. Distill

These models can then be moved safely across company, legal and compliance boundaries.

3. Condense

Before then being used to generate statistically equivalent synthetic data.

4. Use

That's drop-in compatible with your existing analytics code and workflows.

With privacy you control

Hazy generates smart, sample based synthetic data that's differentially private by design. Watch the video below to see how it works:

Enterprise component architecture

Hazy software is divided into three primary components: the synthesiser, hub and client. These are packaged using Docker and can be easily composed by enterprise orchestration systems like Kubernetes and OpenShift. The Hazy synthesiser is designed to work without network access and within air-gapped deployment environments.

Access control, privacy controlsSecurity approvederrors/eventsProduction dataConfigurationReadLive data environmentPush generatorPull generatorClientSynthetic dataLab/ cloud/ test environmentUser credentialsWriteGovernanceMonitoringSecure subnet Generator registryContains read data adapters and the code and algorithms to train and evaluate generative models.Writes out serialisable artefacts and works without network access1. SynthesiserWeb-based management interface, control plane and generator registry. Typically deployed in less sensitive but still private environments.Role-based access control (RBAC).2. HubEnd user client library to pull down and work with generators.Can be installed in any environment.3. Client
Tom Davis
The integrity you can achieve with Hazy’s synthetic data is just amazing.
Tom Davis
Next step

Want to do more with your data?

Hazy unlocks the real value of your data. Without compromising privacy.

Contact us