AUTHENTICATION

TRAC plugs into your enterprise-wide SSO mechanism for watertight identity and access management. Roles can be assigned in the AD for maximum security, with fine-grain permissions configurable in TRAC’s Policy Service.

MODEL REPOSITORIES

As standard, TRAC integrates with GitHub, GitLab, Nexus and BitBucket but custom repo integrations are also possible.

DEVELOPER TOOLS AND IDEs

Deploy the TRAC (Python) runtime library to your IDE and your developers can enjoy all of the type safety and validation of production and build models which can be deployed with a single click.

DATA LAYER

TRAC uses regular cloud buckets (Athena, Big Query or ADF) for scalable, resilient and cost-effective storage. Explore your data with standard SQL and TRAC translates your queries into your cloud platform’s native technologies, giving you the power of cloud through a simple, portable interface. 

EXECUTION SERVICE

TRAC integrates with several execution and compute technologies. For cloud deployments (AWS, GCP, Azure) we provide a Kubernetes pattern as standard supporting both single node and distributed (Spark) capabilities. We also support 3rd party solutions for model execution, as required needed.