THE KUSI WORKFLOW

Your pipeline.
One connected workflow.

Connect data, configure workflows, execute jobs, and inspect results.
Use the interface or Git. Build warehouse models with dbt after data lands.

01
CONNECT YOUR DATA

From source to destination.

Connect databases, warehouses, and survey tools. Choose what to move and where it belongs.

Connections

Manage access to your systems in one place, then reuse connections across workflows.

Managed syncs

Select surveys or database tables and a destination. Kusi creates the sources, tasks, and job.

Fine-grained control

Configure individual sources, destinations, and tasks when a workflow needs more detail.

Explore connection guides
02
CONFIGURE YOUR WORKFLOW

Visual setup. Versioned definitions.

Set up workflows in the UI or declare configuration in YAML. Use schedules and task dependencies to coordinate the work, then build dbt models in your warehouse after jobs succeed.

Same workflow. Your choice.
Configure a syncInteractive example
01 / CONFIGURE

Timezone: UTC

02 / YOUR WORKFLOW

Nightly orders sync

Existing tasks in this job

load.orders
load.customers
Also available as a YAML definition
Explore the same job through either view. Local example; nothing is saved or executed.
SYNC AS CODE

A sync definition
you can review.

Declare sources, destinations, tasks, and jobs in YAML. Keep SQL beside your configuration, and review changes in the same repository as the rest of your work.

Version the whole workflow

Source and destination references, transfer modes, schedules, and task dependencies belong in your manifests.

Plan before you apply

A push produces a plan. Inspect what will change, then explicitly apply the reviewed configuration.

Keep operational control

Pause, resume, and run jobs from the interface. Configuration stays in Git; day-to-day operation stays accessible.

Read the configuration guide
03
ON-DEMAND COMPUTEOperator-enabled rollout

Spin up for the sync.
Wind down after it.

Eligible organizations can select a granted compute profile on a job or Managed Sync. Kusi provisions a worker, runs the job, and releases capacity when its lease closes. Operators enable the isolated runtime for each rollout cohort.

Choose an execution profile

Describe the capacity the job needs through a named compute profile.

Declare runtime policy

Keep concurrency and idle-timeout settings with the job configuration.

Follow the lifecycle

Make provisioning, readiness, and teardown visible alongside execution.

Compute architectureIllustrative · Limited rollout
YOUR WORKLOAD. YOUR CAPACITY.

More work. More workers.

Explore a pool of machines running jobs in parallel.

16
Your configurationExecution profile · concurrency policy
Job queueWork waiting for capacity
PROVISION & DISTRIBUTE
WORKER POOL3 machines · illustrative capacity
machine-01
Worker 01
JOBsync-01
machine-02
Worker 02
JOBsync-02
machine-03
Worker 03
JOBsync-03
One worker per machine in this illustrationScale out →
Work completes Release capacity
Measured usage CreditsInteractive architecture concept · dedicated compute requires operator enablement.

Dedicated execution requires an operator-enabled rollout and a granted profile. Shared workers remain available to all workspaces.

04
DATA OBSERVABILITY

See what ran.
Understand what happened.

Bring execution monitoring and data quality signals into the same workflow. Check destination freshness and row volumes, review reported validation issues, and trace failures back to their tasks.

Monitor freshness & volume

See when destinations last received data, how many rows were written, and row-count trends where tracking is enabled.

Inspect quality signals

Review validation errors, transformation errors, and data inconsistencies recorded in task logs.

Investigate with context

Follow execution status and timing, then use task logs, run history, and alerts to investigate failures.

Destination healthIllustrative data · 7 runs
WAREHOUSE / ORDERS

Know when the data changes.

Last write · Sep 27, 03:00 UTC
Rows written per runLast 7 runs
30k15k0

Select a run to inspect its volume and reported issues.

SELECTED RUNSep 25 · 03:00 UTC

nightly.orders

Rows written
8,120
Duration
0m 38s
Execution
Completed
3 reported issues

Validation warnings recorded by load.orders. Inspect the task log alongside this change in volume.

load.orders3 validation warnings
Volume, freshness, and task logs together. Select a run to explore this illustrative data.

Quality reporting combines task completion metrics with validation, transformation, and consistency issues recorded in task logs.

04
CREDIT-BASED USAGE

Know what your
workflows consume.

Kusi records usage in credits, with visibility into individual job runs and your organization's balance. The pricing direction is simple: pay for the work you use.

Explore the credit model

Usage, measured

Shared runs account for rows, compute time, and data transferred. Dedicated leases are metered by their profile rate.

Run-level visibility

See credit consumption in job history and review usage over a date range.

One organization balance

Track credits added, credits consumed, and the balance available to your team.

05 / OPERATE

Everything around the sync.

The supporting capabilities that keep real workflows manageable.

Connections & data

Sources, destinations, managed syncs, and data exploration.

Scheduling & dependencies

Cron schedules, reusable tasks, transfer modes, and dependency graphs.

Monitoring & diagnosis

Run history, progress, execution logs, alerts, and performance metrics.

Access & extensibility

Role-based access, audit logs, SQL Lab, and API access.

06
ENTERPRISE DIRECTION Planned

More teams.
More control.

Our next priorities build on the controls already in Kusi. These additions are planned; scope and availability will evolve as we work with teams.

Predictable spending

First priorities

TodayRun-level credit usage and organization balances.

Explore planned additions
In the interface
Pre-run estimates, estimated vs. actual usage, and spending alerts.
Behind the workflow
Job and team budgets, server-rate accounting, and usage exports.

Identity & access

First priorities

TodayOrganization roles, API keys, and audit logging.

Explore planned additions
In the interface
Central access administration and clear effective permissions.
Behind the workflow
SSO, SCIM provisioning, and granular resource permissions.

Controlled releases

First priorities

TodayGit definitions, configuration plans, and explicit apply.

Explore planned additions
In the interface
Environment labels, readable diffs, and approval screens.
Behind the workflow
Approval gates, environment promotion, and configuration rollback.

Operations at scale

Planned

TodayJobs, connections, and run history in one application.

Explore planned additions
In the interface
Global search, saved views, ownership labels, and bulk actions.
Behind the workflow
Team-scoped resources, service accounts, and quotas.

Data reliability

Planned

TodayFreshness, row-volume signals, and reported task-log issues.

Explore planned additions
In the interface
Schema-change previews, lineage views, and actionable diagnostics.
Behind the workflow
Schema policies, explicit quality checks, and controlled backfills.

Enterprise trust

Planned

TodayAudit records and documented deployment configuration.

Explore planned additions
In the interface
Security and deployment information, status, and incident history.
Behind the workflow
Residency controls, external secrets, audit export, and documented recovery and service commitments.

Bring a workflow. Make it yours.

Explore Kusi with your own data during the beta.

Request early access