I will build your data pipeline with dbt and BigQuery

Almost every project I take on starts the same way: two people pull the same number and get different answers. What I build: • Ingestion from your sources — database, Stripe, the CRM, whatever exports you are emailing around today • A modelled warehouse in dbt: staging, intermediate and mart layers, each with a job you can explain • Tests on the things that matter — uniqueness, nulls where there should not be any, totals that must reconcile • Documentation generated from the models, so a definition lives next to the code rather than in someone's head • Scheduling with alerting, so a failed run tells somebody instead of quietly serving yesterday's numbers I will tell you which metrics are genuinely ambiguous. Half the disagreements I am asked to fix are not technical — two teams mean different things by "active customer" and nobody has written it down.