Enterprise Data Warehouse Migration to Google BigQuery
80 TB warehouse moved from SQL Server and a 15-node Cloudera cluster to BigQuery: $550K annual savings, 75% lower infrastructure cost, 7x faster workflows.
Case study$550Kannual savings
Ahmet KalafatI make data ready for decisions and for AI.
I build reliable, traceable data platforms so analytics teams and GenAI projects start from data they can trust.
- 01SourceRaw, scattered, many formats
- 02FlowIntegration and orchestration
- 03ModelQuality, meaning, governance
- 04Decision · GenAIDashboards, APIs, RAG
What I do
Three areas, one question: can we trust this data enough to decide?
Reliable data
Layered warehouse design, dimensional modelling and versioned ELT with CI/CD and quality assertions across 3,000+ tables, plus catalog and lineage across 100+ sources.
Decisions
Analytics platforms people can trust and afford: an 80 TB warehouse moved to BigQuery at 75% lower infrastructure cost, and GA4 metrics rebuilt in one Gold layer.
AI
LLM agents for data onboarding, checked by evaluation frameworks (golden sets, trajectory, prompt injection, drift), while production stays deterministic and auditable.
Featured projects
Every project is judged the same way: how close did the data get to the decision it serves?
GA4 Analytics Modernization: Gold Layer, Streaming and Anomaly Detection
GA4 analytics rebuilt on BigQuery with a Gold e-commerce model and drift checks: monthly cost $9K to $3K, heaviest queries 46 minutes to 8 seconds.
Case studyContract-Driven Data Platform with GenAI-Assisted Onboarding
New data sources join through validated YAML contracts; LLM agents work at design time only, so production runs stay deterministic.
Case studySelected publications
Articles and talks: what I learn in practice, written down.
About
The short version. The long version, the timeline and the skill groups are on the About page.
25 years, from software engineering into data warehousing, big data and cloud.
Data Solution Architect on Google Cloud at LC Waikiki. I designed and led the migration of its 80 TB enterprise data warehouse to BigQuery, and today I set the target architecture, the modelling and quality standards and the reference implementations for its data platform.
- Now
- Data Solution Architect, LC Waikiki
- Focus
- Streaming on Google Cloud, GenAI-enabled data products, FinOps at scale
- Education
- MBA, Financial Economics, Marmara University
- Languages
- English, Turkish
Let’s talk.
Data platform architecture, BigQuery migration and FinOps, data readiness for GenAI and AI governance, or an invitation to speak: email me.