Skip to content
Rime Documentation

Guided Onboarding

When you create a new project in Rime, the guided onboarding wizard helps you through the initial setup. It presents a checklist of steps that take you from an empty project to a working data pipeline with monitoring enabled. The checklist tracks your progress and persists across sessions, so you can work through it over multiple sittings.

The onboarding checklist

The wizard covers six steps. Each step becomes available once the previous step is complete, though you can skip steps that do not apply to your situation.

Step 1: Connect Snowflake

Provide your Snowflake account details so Rime can manage your data warehouse. You need:

  • Snowflake account identifier
  • Authentication credentials (username/password or key pair)
  • A role with appropriate permissions

This step is required — Rime needs a Snowflake connection to function. All other steps depend on it.

See Snowflake Resources for details on connection configuration and required permissions.

Step 2: Import or create infrastructure

This step branches depending on whether you are starting fresh or migrating an existing stack:

Existing stack: Use the import tools to bring your current resources under Rime management:

New stack: Create infrastructure resources directly through Rime’s UI:

  • Create databases and schemas for your raw, staging, and mart layers
  • Set up warehouses sized for your workload
  • Configure roles and grants following Rime’s recommended access model
  • Provision S3 buckets and IAM roles for the extraction pipeline

You can also mix the two approaches — import some resources and create others.

This step is marked complete when at least one infrastructure resource (Snowflake or AWS) exists in the project.

Step 3: Configure your first connector

Set up a connector to your first data source. The 5-step connector wizard walks you through selecting a source type, entering connection details, testing the connection, selecting tables, and setting a schedule.

If you are migrating from Fivetran, Stitch, or another extraction tool, configure Rime connectors for the same sources. You can run both tools in parallel during the transition.

This step is marked complete when at least one connector is saved (it does not need to have run yet).

Step 4: Run your first sync

Trigger a manual sync of the connector you just configured. This validates the full extraction pipeline:

  1. Rime connects to the source system
  2. Extracts data as Arrow record batches
  3. Writes Parquet files to S3
  4. Snowpipe loads the data into raw tables in Snowflake

If the sync succeeds, you have a working extraction pipeline. If it fails, the run detail page shows what went wrong (authentication, network, permissions, schema mismatch) so you can correct the issue and retry.

This step is marked complete when at least one connector run finishes successfully.

Step 5: Set up transformation

Configure your first transformation project to turn raw data into something useful:

Existing dbt project: Import your dbt project from Git to bring existing models under Rime management.

New transformation: Create a transformation project in Rime’s UI:

  1. Select a methodology (Kimball or Data Vault)
  2. Choose source tables from your raw layer
  3. Configure staging models (renaming, type casting, filtering)
  4. Define dimension and fact tables (Kimball) or hubs, links, and satellites (Data Vault)
  5. Build mart tables for business consumption

This step is marked complete when at least one transformation project exists with at least one model.

Step 6: Enable monitoring

Set up monitoring to stay informed about your pipeline’s health:

  1. Review the default alert rules that Rime creates for new projects
  2. Configure at least one notification channel (email at minimum)
  3. Optionally set up escalation policies for critical alerts
  4. Review the monitoring dashboard to confirm metrics are being collected

This step is marked complete when at least one notification channel is configured.

Progress tracking

The onboarding checklist appears as a progress bar at the top of the project dashboard until all steps are complete (or dismissed). Each step shows one of three states:

  • Incomplete — the step has not been started or its completion criteria are not met
  • Complete — the step’s criteria are met
  • Skipped — you chose to skip this step

Your progress is saved to your project and persists across browser sessions and across team members. Any team member can continue where another left off.

Skipping steps

Not all steps are mandatory. You can skip any step except Step 1 (Connect Snowflake). Reasons you might skip:

  • Skip Step 2 if you plan to create infrastructure later or if your Snowflake account is already configured and you do not need Rime to manage the objects.
  • Skip Step 3 and 4 if you are only using Rime for transformation and monitoring of data that is already in Snowflake via other tools.
  • Skip Step 5 if you do not need transformations yet (e.g., you are starting with extraction and monitoring only).
  • Skip Step 6 if you want to configure monitoring later.

Skipping a step does not prevent you from completing it later. You can return to any skipped step from the checklist at any time.

Greenfield (no existing stack)

Follow the steps in order: connect Snowflake, create infrastructure, add connectors, run syncs, build transformations, enable monitoring. This is the default flow and the one the wizard is optimized for.

Existing Snowflake + dbt

  1. Connect Snowflake (Step 1)
  2. Import from Snowflake (Step 2)
  3. Import dbt project (Step 5)
  4. Configure monitoring (Step 6)
  5. Add connectors later as you migrate off your existing extraction tools (Steps 3-4)

Existing Snowflake + extraction tools, no dbt

  1. Connect Snowflake (Step 1)
  2. Import from Snowflake (Step 2)
  3. Configure connectors to replace your existing extraction tools (Steps 3-4)
  4. Build transformations (Step 5)
  5. Enable monitoring (Step 6)

Monitoring only

If you want to use Rime purely for monitoring an existing stack before migrating management:

  1. Connect Snowflake (Step 1)
  2. Skip Steps 2-5
  3. Enable monitoring (Step 6)

You can return to the other steps when you are ready to bring more of your stack under Rime management.

Dismissing the checklist

Once you have completed or skipped all steps, the checklist disappears from the project dashboard. You can also dismiss it early by clicking Dismiss onboarding in the checklist menu. Dismissing does not affect your project configuration — it only hides the checklist UI.

If you want to see the checklist again after dismissing it, go to Project > Settings > Onboarding and click Show checklist.

Next steps