Cloud & Data Engineering
Google Cloud, BigQuery, and Vertex AI pipelines that ship.
Data pipelines and ML workflows on Google Cloud — BigQuery warehouses, Vertex AI training and serving, cost-efficient for teams beyond spreadsheets.
Problems I solve
- Data locked in spreadsheets and disconnected apps
- Ad-hoc reports built by hand every week
- No warehouse, no single source of truth
- ML models built but never shipped
What you get
- BigQuery warehouse with modeled tables
- Ingestion pipelines from apps, ads, and CRMs
- Vertex AI training and serving
- Dashboards in Power BI / Looker Studio
Use cases
Marketing analytics warehouse
Unify GA4, Meta Ads, Google Ads, HubSpot, Stripe into one BigQuery source of truth.
Product analytics on BigQuery
Event pipeline + dbt models + Looker Studio dashboards for PLG metrics.
ML feature store lite
Feature tables in BigQuery + Vertex AI training + online serving via Cloud Run.
Cost & FinOps dashboards
Attribute LLM, cloud, and SaaS costs per feature / customer / team.
Examples I've shipped
DTC brand — unified analytics
Outcome — Marketing ROAS visible per campaign per SKU, refreshed hourly.
Churn prediction on Vertex AI
Outcome — 12% reduction in monthly churn via targeted saves.
Case study
Challenge
Founder pulling reports manually from Shopify, Meta, Klaviyo every Monday, no view of blended ROAS or LTV.
Approach
- Fivetran ingestion into BigQuery for 7 sources
- dbt models for orders, sessions, and marketing spend
- Looker Studio dashboards for exec + marketing
- Weekly automated Slack digest with anomalies
Results
How I work
- 1
Audit
Map current data sources, gaps, and reporting needs.
- 2
Warehouse design
Model tables, choose partitioning, plan cost.
- 3
Ingestion
Build pipelines from source systems to BigQuery.
- 4
Modeling
Transform raw data into analytics-ready marts.
- 5
Dashboards
Ship decision-ready views for stakeholders.
Deliverables
- BigQuery warehouse with documented schema
- Ingestion pipelines
- Analytics-ready data models
- Stakeholder dashboards
Benefits
- One source of truth for the business
- Reports that build themselves
- Foundation for ML and AI features
Frequently asked questions
Why BigQuery over Snowflake or Redshift?+
BigQuery has zero infra to manage, transparent per-query pricing, and native ML — the right default for most SMBs I work with.
Can you connect our current tools?+
Yes — HubSpot, Shopify, Google Ads, Meta Ads, Stripe, and custom sources via Airbyte, Fivetran, or Python.