Run your data warehouse end-to-end.
Best-practice guides organized by role. Each one teaches the practice first, then shows how Kirimana operationalizes it. Pick your role to see the track built for you.
The catalogue is open; reading a guide requires a (free) account. Sign in or request access .
11 guides in this track · updated with every release
Strategy & operating model
- Design a data governance operating model that scales 🔒
Hub-and-spoke governance where a central team sets the rails and domains own their data, enforced as platform rules rather than review meetings.
- Make governance a property of the platform, not a committee 🔒
Governance that executes at pull-request time beats policy documents nobody reads. The platform refuses violations before a reviewer ever has to catch them.
- Choose a platform strategy that survives re-orgs 🔒
Platform-agnostic contracts keep your governance and metadata portable across runtimes, so a cloud migration or a re-org does not force you to rebuild your data platform from scratch.
Data governance
- Set an AI governance posture the board can defend 🔒
Move artificial intelligence (AI) governance from a slide deck to an enforced gate: a per-dataset AI policy, an autonomy ladder with a hard ceiling, and an audit record behind every model call.
- Run one governance model across every platform 🔒
Owners, classification, lineage, AI usage, and audit should follow a single contract model regardless of runtime, so the whole estate stays auditable instead of fracturing into per-platform silos.
Compliance & audit
- Generate compliance evidence instead of promising it 🔒
Produce a structured evidence document from contract metadata and the audit log over a defined control set, machine-readable for governance tooling and human-readable for reviewers, with gaps flagged rather than hidden.
- Answer an audit without a six-week investigation 🔒
When the auditor asks how a dataset became restricted, who approved it, and where it flows, the answer should be a link rather than a task force, because traceability was designed in instead of reconstructed.
Value & adoption
- Prove data platform value in weeks, not quarters 🔒
Ship one source through to one trusted number before you scale out. Time-to-first-value is the metric that protects your budget when the board asks what the platform has delivered.
- Trace every metric back to its sources 🔒
When the board asks where a revenue number comes from, answer in one query rather than a six-week investigation. Goal-to-data lineage turns traceability into a trust instrument.
- Attribute data platform cost to the work that caused it 🔒
FinOps for data. Trace every unit of spend back to the dataset that generated it, so cost becomes a governance signal that tells you what to retire rather than a mystery bill.
- Measure and grow your data and AI maturity 🔒
You cannot improve what you do not measure. Baseline your maturity, invest where the gap is widest, and make every headline metric cite the sources it comes from.