How Block Deployed AI Agents Company-Wide in 2 Months

What happens when a single engineer’s side project turns into a company-wide AI platform used by every department—engineering, product, marketing, finance, customer support, and sales?

In this episode of the Hangar DX Podcast, Block’s VP of Engineering Angie Jones shares the inside story of how Block deployed AI agents across the entire organization in just eight weeks.

She reveals how an internal tool called Goose—originally built by one engineer—became one of the first-ever MCP clients, exploded as an open-source project, and evolved into a general-purpose agent powering workflows across the company.

We also dig into security and governance, adoption strategies, and practical lessons that every platform team can learn from.

00:00 Introduction to AI Transformation at Block
04:18 Building Goose: The AI Agent
08:50 Adoption Across Departments
11:59 Scaling MCP Servers
13:17 Technical Challenges with MCPs
16:33 Governance and Security of MCPs
17:35 Tool Overload and Centralization Strategies
20:09 Overcoming Cold Start Problems
23:34 Evolving Goose for Different Departments
25:58 Open-Source and Internal Development
28:05 Measuring Success in AI Initiatives
30:02 Recommendations for AI Adoption
33:19 Future Predictions and Initiatives

About Angie Jones
Angie Jones is the Vice President of Engineering, AI Tools & Enablement at Block, Inc.

She is an award-winning teacher and international keynote speaker and holds more than 25 patents for inventions in the areas of virtual worlds, collaboration software, social networking, smarter planet, and software development processes.

About Hangar DX (https://dx.community/)
The Hangar is a community of senior DevOps and senior software engineers focused on developer experience. This is a space where vetted, experienced professionals can exchange ideas, share hard-earned wisdom, troubleshoot issues, and ultimately help each other in their projects and careers