data.gm.com Learning Week - September 22-24

Keynote

How to use #help-data to Increase Productivity + Serval Demo

Use the #help-data channel as your central place to ask any data or Databricks questions so blockers are removed quickly and you can keep work moving. Serval tool Demo

Panel: Trusted Data Starts with Stewardship

Google Workspace Demo

Discover how Google Workspace will power the next chapter of collaboration at GM. In this live demo, we’ll walk through key Workspace tools showing how they work together as one connected platform. Whether you’re curious about what’s changing from Microsoft 365 or just want to see what’s possible, you’ll leave with a clear picture of how everyday work will feel in Google Workspace and where to go for training and support.

Self-driving Data Quality in GM - a quick peek at Anomalo

Anomalo is a data quality software platform that helps domains automatically monitor, detect, and resolve issues in their data. This session provides an overview of Anomalo's core capabilities, including automated anomaly detection, data validation, and root cause analysis across your data warehouse. We'll also explore Anomalo's agentic functionality — intelligent, self-driving automation that continuously monitors your data and helps ensure your organization can rely on trusted, high-quality data without manual rule-writing or constant oversight.

Using AI Paired Programming to Deliver Power BI Reports and Dashboards

Showcase how GitHub Copilot CLI and Fabric Skills can assist Power BI developers to rapidly redesign, optimize, and enhance reports.

What is a Context Layer? How GM is using Atlan to solve this.

Everyone is racing to ship AI agents; few can trust the answers. This session shows how GM can use Atlan’s Context Engineering Studio and Context Agents to build a governed context layer that sits across warehouses, BI, and operational systems, then feeds consistent, testable context into any agent. We’ll walk through how AI bootstraps most of the semantic model from GM’s existing dashboards and SQL, how domain experts capture the last-mile business logic, and how evaluation suites and traces validate agent behavior before rollout. Attendees will leave with a concrete blueprint for engineering GM’s context layer so agents answer like GM does—not like a generic LLM.

Building Connected AI Agents with Azure AI Foundry

Learn how Azure AI Foundry enables organizations to build and govern AI agents that work with existing platforms such as Azure Databricks, Glean, and other enterprise applications.

Data Contracts

This presentation explains how data products and data contracts work together to deliver reliable, reusable data. We’ll clarify what makes a good data product, break down the key parts of a data contract (schema, data quality, SLAs, ownership, and governance), and show practical examples of how contracts set clear expectations between data producers and consumers. You’ll leave with a straightforward understanding of how to design, document, and govern data products using contracts so they are easier to find, trust, and use across the organization.

Data Governance & AI Governance

Learn the fundamentals of data and AI governance, including data quality, stewardship, compliance, and responsible AI practices that help ensure trusted data and secure, ethical AI outcomes.

ROAD - Why ROAD?

Learn how to use ROAD’s built-in ‘slicing’ feature to put record-series retention requirements into practice. In this 30-minute session, we’ll walk through how to identify the appropriate trigger date for a record series, use ROAD filters to categorize records as either eligible for disposal or required for continued retention, and organize eligible records into manageable annual batches. Participants will gain practical guidance for partnering across business and technical teams to review, validate, and take action on retention and disposal activities in a structured, repeatable way.

Genie, Glean & Agentic AI

What we've done - where we're going?

Data Transfer Tools Demo

Learn how GM securely moves data across teams and partners through our External Data Transfer and Inbound Data eXchange processes. We’ll break down real scenarios, the governance controls that keep us compliant by design, and the automations that cut risk while speeding collaboration. Join us to learn practical ways to streamline data exchanges, protect sensitive information, and keep GM secure.

Find It. Request It. Use It: Self-Service Data Access in Atlan

Join us for a practical walkthrough of GM’s streamlined data discovery and access process in Atlan. We’ll demonstrate how to search for the data you need, review asset details and context, and submit an access request directly from the catalog. You’ll see how Atlan serves as the starting point for discovering schemas, tables, and data products, while routing governed access requests to the appropriate approval workflow. The session will also cover what information to provide in a request and how this process supports faster, more transparent, and auditable access to data

Genie workshop with Databricks

Genie Agents are domain-specific environments where you configure trusted data, metrics, and business rules to power natural-language analytics. Genie Agents let people ask questions of your data in plain language and get real answers back. Whether it works well depends almost entirely on how the underlying data is described and curated, which is the part most teams skip. This is a working session, not a demo. Every participant builds their own space, teaches it the language of their business, then tests it honestly and sees which questions it gets wrong. For analysts and practitioners who will own and maintain a space, not just use one. Requirements: Please bring your sandbox environment and also be prepared to destroy the space once done.

AI on Databricks (advanced)

In this session we will walk through the process of building, deploying, and monitoring production-grade custom agents on Databricks.

Vibecoding Apps (technical + advanced)

AI coding assistants can now produce a working internal app in minutes. The more interesting question is what has to be true before that app touches real data. We build and deploy a real app, from a blank slate to a record landing in a table, then walk through exactly what we would check before letting it run in a production environment. This is for anyone who builds internal tools, dashboards or data entry apps. Python or SQL helps but is not required.

The Human Aspect of AI

Why healthy AI adoption builds judgement instead of eroding it.