Setup

Rebtel-specific setup on top of the shared Development Environment.

Data Warehouse Connection

Connect your dbt development environment to Snowflake with your own user.

dbt Core (recommended)

Use SSO in ~/.dbt/profiles.yml. No password or MFA setup needed:

authenticator: externalbrowser

To use username and password instead, set up MFA with the Duo app and use:

authenticator: username_password_mfa

See the dbt Snowflake setup docs.

dbt Cloud IDE & CLI

1. Go to Account Settings > Your Profile > Credentials > Data Team's DBT Project
2. Under Development credentials, choose Username and password and fill in your details

Note

Password login requires MFA with the Duo app.

VPN

Optional, only needed to connect to the BE DBs directly. Reaches services in Rebtel's dedicated Azure tenant.

1. Get a rebtel.io Microsoft account from DevOps and access to the Azure Portal
2. Install Azure VPN Client (Mac users need an Apple account for the App Store)
3. Import the VPN config file into Azure VPN Client and save it

Repositories

All Data Team repos live in the rebtel GitHub org and are prefixed with data-team-:

• data-team-dbt
  All dbt models built and maintained by the Data Team, see dbt Project Structure
• data-team-looker
  All LookML code in the rebtel_production project in Looker, see Looker
• data-team-snowflake-objects
  Record of every Snowflake object the Data Team owns: pipes, tasks, alerts, stages, procedures, integrations and policies. Snowflake is the source of truth and changes are mirrored here, see Snowflake
• data-team-lambda
  All AWS Lambda functions of the Data Team, see Architecture Overview
• data-team-rebtel-data-warehouse-mcp
  Semantic layer served over MCP: a curated catalog of the warehouse (glossary, joins, domains, references, SQL exemplars) that agents check before writing SQL
• data-team-rebtel-viz
  Rebtel Viz, chat-published charts and dashboards on Snowflake at viz.reblab.net. Port of Majority's viz app to Vercel and Neon
• data-team-streamlit-app
  Streamlit in Snowflake apps
• data-team-view-order-shap-ui
  SHAP viewer: type an order ID to see the PROD and SHADOW fraud model explanation images, also served as a JSON API
• data-team-risk-fraud-ml-model
  Production setup of the risk and fraud ML model from GCP JupyterLab