Assess
Review model graph, sources, naming, materializations, and dashboard dependencies.
dbt, SQL, marts, KPI-ready datasets
dbt & Analytics Engineering
For teams whose reporting is slowed down by unclear SQL, fragile dbt/Databricks models, missing tests, or analytics logic spread across dashboards and notebooks.
Services
Make the analytics layer maintainable.
dbt model cleanup
Data mart design
SQL optimization
Tests and documentation
KPI layer / reporting datasets
Dashboard-ready models
Databricks-ready modeling patterns when relevant
Typical cleanup plan
Simplify
Split unclear models, remove duplicated SQL, and define durable marts.
Protect
Add tests, freshness checks, docs, and ownership around critical models.
Ship
Document the model layer and create a practical delivery backlog.
How the engagement works
01
Scope the business problem
Clarify the pipeline, reporting, cost, or delivery pain and decide whether an audit, sprint, or retainer fits.
02
Review the current system
Inspect architecture, data flow, dbt models, AWS services, deployment paths, observability, and cost signals.
03
Ship practical improvements
Implement fixes with Terraform, CI/CD, tests, documentation, dashboards, and handover notes where needed.
04
Hand over clear next steps
Leave the team with recommendations, runbooks, implementation notes, and a prioritized backlog.
FAQ
Are the enterprise examples Riahi IT Solutions client work?
They are founder experience from work delivered through German consultancies before and alongside the formation of Riahi IT Solutions. Public wording is kept NDA-friendly and avoids claiming direct company-client relationships where that would be inaccurate.
What is the best first engagement?
For unclear problems, start with a fixed-scope audit. For a known backlog, use an implementation sprint. For ongoing delivery support, use a part-time consulting retainer.
Do you only work on AWS?
AWS data platforms are the core offer. Adjacent work can include Azure/GCP integrations, Databricks, dbt, SQL, Terraform, CI/CD, Python, reporting datasets, and dashboard-ready data models.
Get a dbt project review.
Send a short description of your dbt project, warehouse, pain points, and reporting goals. Typical response time: within 24 hours.