Bianca Stratulat

What if AI didn’t require Python, notebooks, or ML pipelines? In this session, we’ll use SQL AI Functions (ai_parse, ai_summarize, embeddings, and model calls) to build an AI-powered data pipeline using only SQL on Databricks.

Moving from engineer or analyst to data leader isn’t just a promotion, it’s a complete shift in mindset. As a female leader, completing an MBA, and leading modern data teams, I realised something uncomfortable: no one actually teaches you how to make this transition.

Public sector organisations face increasing pressure to deliver better outcomes despite constrained budgets, complex data estates and increasing regulatory expectations. This session shares how we said yes to an ambitious Data for Good challenge by delivering a production-ready public sector data platform. Rather than treating the initiative as a proof of concept or training exercise, we approached it as a real client engagement, applying delivery standards, governance, accountability and measurable outcomes from day one. Over eight weeks, the team designed and implemented a complete solution, including multi-environment architecture (Dev/Test/Prod), governed data access, automated data pipelines and CI/CD deployment. This session explores the practical decisions that made the project successful: balancing ambition with realistic scope, embedding governance early, managing delivery risk and maintaining production-quality standards throughout. It also demonstrates how delivering a real project accelerates capability while creating tangible public value. Attendees will leave with a practical, repeatable framework for planning and delivering Data for Good initiatives that are technically robust, commercially viable and ready for production. All on Databricks.

SQL didn’t get weaker, data platforms got better. In this session, we demonstrate how modern data engineering on Databricks can be delivered end-to-end using SQL alone on the Lakehouse, without dependency on Python or external frameworks, including SQL AI functions and Geospatial Functions. Designed for organisations with strong BI/SQL foundations, this session shows how to transition to production-grade data engineering with minimal upskilling. We showcase how pipelines are built using Lakeflow Spark Declarative Pipelines via the Pipeline Editor, with ingestion, transformation, and data quality rules all defined declaratively in SQL. This demonstrates how SQL moves beyond querying into full pipeline engineering. Governance is implemented by design using Unity Catalog, providing fine-grained permissions, lineage, and auditability across the entire pipeline. We also showcase AI/BI Genie, configured using SQL-only approaches (metrics, dimensions, instructions, and evaluation queries), enabling natural language access to trusted data. This session reframes SQL as a first-class engineering language on the Lakehouse, capable of handling streaming, transformation, data quality,…

Everyone talks about Fabric and Databricks together, few make it work in production. This session shares real-world patterns for integrating Databricks, Fabric and Power BI, covering semantics, performance, CI/CD, governance, and cost trade-offs beyond the PoC.

NHS England is revolutionizing healthcare research by enabling secure, seamless access to de-identified patient data through the Federated Data Platform (FDP). Despite vast data resources spread across regional and national systems, analysts struggle with fragmented, inconsistent datasets. Enter Databricks: powering a unified, virtual data lake with Unity Catalog at its core — integrating diverse NHS systems while ensuring compliance and security. By bridging AWS and Azure environments with a private exchange and leveraging the Iceberg connector to interface with Palantir, analysts gain scalable, reliable and governed access to vital healthcare data. This talk explores how this innovative architecture is driving actionable insights, accelerating research and ultimately improving patient outcomes.