A practical FinOps playbook for Databricks. Cluster types, file compaction, caching, serverless, and BI rationalization — with realistic savings ranges.
Power Bi Direct Lake on Databricks SQL: a Modernization Playbook
Migrate Power BI semantic models from import / DirectQuery to Direct Lake on Databricks SQL. Performance, governance, and migration patterns.
Production ML on Databricks: Mlflow, Feature Store, Calibration
A reference MLOps stack on Databricks — MLflow Model Registry, Feature Store with online serving, calibration-first model evaluation, and Mosaic AI Model Serving.
Building Multi-Agent Workflows on Databricks (mosaic AI Agent Framework)
Multi-agent workflows native to the Lakehouse — designed, built, evaluated, and deployed on the Mosaic AI Agent Framework with typed tools and an evaluation harness.
Production-Grade RAG on the Lakehouse with Mosaic AI Vector Search
How to design, build, and evaluate a production RAG system on Databricks using Mosaic AI Vector Search, hybrid retrieval, and a real evaluation harness.
Unity Catalog: Governance Done Right
Most governance projects fail because they start with policy. The good ones start with structure. Here is a reference Unity Catalog deployment that supports both governance and data-mesh patterns.
Streaming on the Lakehouse: Auto Loader + DLT in Practice
A reference architecture for real-time pipelines on Databricks. Auto Loader, DLT, expectations, and SLOs that survive production.
Dimensional Modeling on Delta Lake — and When to Choose Data Vault Instead
Star schema, data vault, one-big-table, federation. Each has a different shape and different trade-offs on Delta Lake. A decision framework.
Modernizing ETL: Informatica/ssis/datastage to Lakeflow + DLT
A practical conversion playbook for legacy ETL — Informatica, SSIS, DataStage — to Databricks Lakeflow + DLT with Auto Loader. Patterns, expectations, and how to handle SCD types.
OBIEE to Databricks: a Practical Migration Pattern
Move Oracle OBIEE / OAS to Databricks SQL with a clear semantic-layer methodology. RPD reconstruction, security translation, ETL conversion, and report rebuild — without losing business logic.