Building an AI Ready Finance Lakehouse: dbt Models, Governed Metrics, and Agentic Operations on Data

Large finance organizations rely on dashboards and reports, but the harder challenge is enabling automation without breaking governance or auditability. This session shows how a global finance organization built agent-driven workflows on the Databricks Data Intelligence Platform using dbt for analytics engineering, Unity Catalog for governance, and certified metrics views.

We walk through a real implementation where transformations are standardized through dbt models across staging, intermediate, and marts layers using modular SQL, macros, and reusable patterns. dbt tests, documentation, and exposures enforce data quality, lineage, and traceability from source to finance-ready outputs. Metrics are defined once, protected with attribute-based access control and automated classification, and reused across reporting and analytics.

Power BI connects to governed lakehouse tables, while Iceberg-compatible engines access Delta tables through Uniform for reuse.

AI-Powered Real-Time Intelligence: Transforming Data-Driven Decisions with Microsoft Fabric & GenAI

In today’s fast-paced world, making smart, data-driven decisions in real time is essential for maintaining efficiency and quality in business operations. Join us to explore Microsoft Fabric Real-Time Intelligence and discover how to seamlessly ingest, analyze, and act on real-time data.
Learn how to:
1. Ingest real-time data from multiple sources into Fabric
2. Run lightning-fast queries for instant insights
3. Monitor key metrics with intuitive, real-time dashboards
4. Set up automated alerts and detect anomalies for proactive issue resolution
5. Leverage Copilot in Real-Time Intelligence to query data using natural language

Orchestrating Multi-Agent AI with Fabric RTI: From Alert to Action

Wire Fabric RTI end-to-end: Eventstream -> Eventhouse (KQL) flags anomalies; Activator (GA) fires; Copilot Studio agents + computer use (preview) consult a Digital Twin (preview) and execute safe remediations - governed on OneLake with Data Agents.

Oracle↔Lakehouse: Real-Time CDC to Databricks, Served Back to Oracle

Make Oracle the hero at both ends of your analytics pipeline. In this session, we start with Oracle Database as system of record and stream real-time CDC using Oracle GoldenGate Microservices into object storage/Kafka. Databricks ingests changes incrementally with Auto Loader, applies governed transformations (Unity Catalog permissions, lineage, quality checks), and merges safely into Delta Lake (bronze→silver→gold). Finally, we serve curated data back to Oracle so existing Oracle Analytics Cloud (OAC) dashboards and SQL tools consume near-real-time, trusted tables—no dashboard re-platforming required.
You’ll see resilient CDC patterns (late/duplicate events, PK updates), schema evolution and SCD handling, merge best practices, and observability/SLA tips. The live demo walks the full loop: Oracle DB → GoldenGate → object storage/Kafka → Databricks (Auto Loader, MERGE, lineage) → Oracle destination → OAC visualization. Leave with an architecture, checklists, and code snippets you can productionize.

From Signal to Action in Fabric RTI: Data Agents, Digital Twins, Activator & Copilot

Operations don’t improve because we see anomalies—they improve when we act on them safely and repeatably. This session shows how to design a closed-loop “sense → reason → act” system on Microsoft Fabric Real-Time Intelligence (RTI) where Data Agents convert raw telemetry into decisions and auditable actions.
We’ll stream events with Eventstream into Eventhouse (KQL), detect out-of-spec behavior, and trigger Activator reflexes that launch agent plans. Agents consult a Digital Twin (asset models, tolerances, topology) and leverage Copilot Studio (incl. computer use for UI automation) to implement safe remediations—under approvals/guardrails, with full OneLake lineage and audit.
The talk is vendor-neutral in patterns and shows interop with common enterprise sources (e.g., Kafka/Confluent, GoldenGate CDC into object storage, REST/DB adapters). You’ll leave with a blueprint—reference architecture, KQL examples, agent tool design, twin modeling tips, approval flows, and an end-to-end demo—that your team can adapt to manufacturing, retail ops, utilities, or SRE use cases to collapse MTTR while keeping humans in control.

Real-Time Intelligence: Turning Streaming Data into Smart Decisions

In today’s fast-paced world, making smart, data-driven decisions in real time is essential for maintaining efficiency and quality in business operations. Join us to explore Microsoft Fabric Real-Time Intelligence and discover how to seamlessly ingest, analyze, and act on real-time data.
Learn how to:
1. Ingest real-time data from multiple sources into Fabric
2. Run lightning-fast queries for instant insights
3. Monitor key metrics with intuitive, real-time dashboards
4. Set up automated alerts and detect anomalies for proactive issue resolution
5. Leverage Copilot in Real-Time Intelligence to query data using natural language

AI-Powered Real-Time Intelligence: Transforming Data-Driven Decisions with Microsoft Fabric & GenAI

In an era where timely, data-driven decisions define competitive success, Microsoft Fabric’s newest real-time intelligence capabilities are setting new standards. This presentation dives into how Fabric’s integrated AI and generative AI (GenAI) features are redefining real-time data analytics and automation. Attendees will explore Fabric’s Real-Time Hub, a streamlined platform enabling organizations to effortlessly bring together diverse data streams, and Event Streams, which seamlessly manage high-volume data capture and distribution from sources.

We’ll examine Eventhouse, Fabric’s optimized engine for high-precision, time-sensitive analytics, alongside Data Activator Integration, which automates alerts and actions triggered by AI-informed, real-time data patterns. With GenAI models embedded in these workflows, users can access predictive insights and dynamic recommendations, allowing for immediate responses to data events.

Additionally, KQL Queryset and Real-Time Dashboards will be featured, highlighting the creation of interactive visualizations, the execution of advanced queries, and the setup of live alerts—all within a single, cohesive framework. Through real-world scenarios and live demonstrations, this session showcases how the synergy of Microsoft Fabric’s AI-powered real-time intelligence and GenAI insights is empowering data teams to make fast, high-impact decisions. This talk will be valuable for data professionals and business leaders looking to leverage cutting-edge AI and real-time intelligence capabilities to drive strategic agility and precision in decision-making.

End-to-End Data Security in Microsoft Fabric: Deep Dive into OneLake Security

Attend this session to explore how Microsoft Fabric’s OneLake Security (preview) enables table- and folder-level security, row-level security (RLS) and column-level security (CLS) inside the OneLake Catalog. We’ll demo creating a security role, assigning access to specific folders/tables, defining filters on rows and columns, and verifying enforcement across Spark notebooks, the SQL analytics endpoint, and Power BI.

Agentic AI in Finance: Governed Metrics, Open Access, and Automated Action on Azure Databricks

Large finance organizations already rely on dashboards and reports, but the real challenge is enabling automation without breaking governance or auditability. This session shares how a global finance organization implemented agent-driven workflows on the Databricks Data Intelligence Platform using Unity Catalog, metrics views, and open lakehouse patterns.

The talk walks through a real implementation where finance metrics are defined once using metrics views, protected with attribute-based access control and automated classification, and reused consistently across reporting, analytics, and automation. Power BI connects directly to governed lakehouse tables and metrics for enterprise reporting, while Iceberg-compliant clients consume Delta tables through Uniform, enabling open access without duplicating data.

Agentic workflows observe certified metrics and events, assist with variance investigation, anomaly triage, and operational follow-ups, and propose actions that remain traceable and reviewable. Early results indicated several hours of manual investigation effort saved per reporting cycle and faster anomaly triage, without introducing additional governance or audit risk. The session includes an architecture walkthrough and demo using synthetic data, reflecting real finance constraints, trade-offs, and lessons learned when moving from reporting to governed, open, and action-oriented analytics.

Designing Conversational BI: How Azure Databricks AI/BI Genie Unlocks Self-Service Insights

In this session, we will explore how Databricks AI/BI Genie enables non-technical business users to interact directly with enterprise data using natural language. You will see how to build a governed data agent that delivers accurate answers, actionable insights, and automatic visualizations, thereby empowering business teams to make faster, data-driven decisions without relying on SQL or dashboards.

Agentic AI in Finance: Governed Metrics, Open Access, and Automated Action on Azure Databricks

Large finance organizations rely on dashboards and reports, but the harder challenge is enabling automation without breaking governance or auditability. This session shows how a global finance organization built agent-driven workflows on the Databricks Data Intelligence Platform using Unity Catalog and metrics views.

The talk covers a real implementation where finance metrics are defined once, protected with attribute-based access control and automated classification, and reused across reporting and analytics. Power BI connects directly to governed lakehouse tables, while Iceberg-compliant engines access Delta tables through Uniform.

Agentic workflows observe certified metrics to assist with variance investigation and anomaly triage. Early results showed several hours of manual investigation effort saved per reporting cycle. The session includes an architecture walkthrough and demo using synthetic data that reflects real finance workflows and design decisions.

From Tables to Agents: Building Enterprise AI Workflows with Azure Databricks Agent Bricks

AI agents are everywhere - but how do they actually work in a real data platform?

In this session, we’ll go beyond chatbots and build a practical, enterprise-ready multi-agent system using Databricks Agent Bricks. You’ll see how structured data (SQL tables, Delta Lake, Genie Space) and unstructured documents (PDFs, invoices, call logs, reviews) can be orchestrated through a supervisor agent to answer complex business questions.

Using a realistic enterprise demo, we’ll walk through:

Extracting invoice data from PDFs and images into structured tables

Grounding answers in product manuals and compliance documents

Analyzing customer sentiment from call transcripts and reviews

Routing tasks across specialized agents using a supervisor

Connecting everything back to SQL analytics and dashboards

This is not theory — this is a working architecture that connects your existing SQL data estate to modern AI workflows.

If you work with SQL, data engineering, or analytics and want to understand how agentic AI fits into the real enterprise stack, this session is for you.

From Data to Decisions: Building Agentic Construction Intel using Azure Databricks Agent Bricks

Construction companies generate massive amounts of operational and financial data across ERP systems, field reports, inspection logs, and project controls tools. Yet most organizations still rely on dashboards and manual investigation to understand project performance.
This session demonstrates how agentic workflows built with Databricks Agent Bricks transform construction data into automated operational intelligence. Using a synthetic construction services scenario, we show how multiple AI agents analyze project financials, field inspections, and operational data stored in the lakehouse.
Specialized agents interpret structured ERP data and unstructured inspection documents, while a supervisor agent orchestrates investigations across multiple sources to detect cost variance, identify recurring defects, and recommend operational actions. The demo illustrates how the Databricks Data Intelligence Platform enables governed, explainable automation while preserving enterprise data governance.
Attendees will learn practical design patterns for building agent-driven systems that move construction analytics beyond dashboards toward actionable intelligence.

This session will be the first public presentation of this demo. The demo uses synthetic construction industry data and does not contain any client or confidential information.

The session includes a live architecture walkthrough and demonstration built on the Databricks Data Intelligence Platform using Agent Bricks, Unity Catalog governance, and lakehouse data patterns.

Target Audience:
Data engineers, data architects, AI engineers, platform engineers, and analytics leaders interested in building enterprise agentic AI systems on governed data platforms.

The session focuses on architecture patterns, governance considerations, and practical design approaches for building multi-agent workflows using enterprise data.