From Systems of Record to Agentic Systems of Action: A MongoDB guide to building agentic AI at enterprise scale

About the Book

Learn how Systems of Action move agentic AI beyond the pilot stage through business context, decision traces, governed action, and coexistence with systems of record Key Features Design an Intelligence Context Layer around enterprise business objects Keep systems of record authoritative while humans and agents act Apply transferable Systems of Action patterns across industries Purchase of the print or Kindle book includes a free PDF eBook Book DescriptionEnterprise AI pilots often stall when agents meet fragmented data, systems designed around human workflows, and production requirements for control and auditability. From Systems of Record to Agentic Systems of Action shows how to build Systems of Action that work with the systems of record an enterprise already depends on, without making wholesale replacement a prerequisite. You'll learn to design an Intelligence Context Layer that keeps business context, enrichments, retrieval structures, and decision traces connected to business objects. The architecture is worked through on MongoDB, showing where agent memory and working state fit, how source systems remain authoritative, and how humans and agents reason and act through governed paths. Written by MongoDB practitioners and industry specialists, the book covers event-driven synchronization, information modeling, semantic retrieval, controlled writeback, multi-agent coordination, human oversight, identity, and auditability. Worked examples span insurance, healthcare, manufacturing, and financial services, each exposing a different operating condition so readers can transfer the architecture to their own industry. MongoDB Vector Search, Time Series, Atlas Stream Processing, MCP, and agentic RAG appear where they solve a concrete design need.What you will learn Keep systems of record authoritative while agents act Design an Intelligence Context Layer around business objects Capture decision traces and decision-relevant context Use retrieval, memory, and state in agent workflows Synchronize context with event-driven data flows Coordinate multi-agent workflows with human checkpoints Design governed writeback, identity, and audit controls Scale agent systems into shared enterprise infrastructure Who this book is forThis book is for enterprise and solution architects, AI architects and engineers, product owners, engineering managers, platform and data engineers, senior developers, data and platform engineers, engineering leaders, and technology leaders responsible for bringing agentic AI into production. Its architecture applies across industries; worked examples come from insurance, healthcare, manufacturing, and financial services. Familiarity with enterprise architecture, data platforms, and core AI concepts is helpful; no prior MongoDB experience is required.

Table of Contents:
Table of Contents

1. From Systems of Record to Systems of Action
2. The Intelligence Context Layer
3. The Coexistence Architecture: How Systems of Action Work with Systems of Record
4. From Experiment to Product: A Mindset and Architecture for Lasting AI
5. Information Architecture for Systems of Action
6. Design and Integration Patterns for Action
7. Bridging Patterns to Practice: How to Read the Use Cases
8. Agent-Enhanced Claim Handling: The Canonical System of Action
9. The Underwriting Intelligence Layer: Parallel Agents and Bounded Authority
10. The Member Care Coordination Layer: From Prior Authorization to Continuous Care
11. A Unified Namespace Architecture for Manufacturing Systems of Action
12. Multi-Agent Systems for Operational Resilience in Manufacturing
13. FSI Document Intelligence with Agentic AI and MongoDB
14. Agentic AI-Powered Investment Portfolio Management
15. Agentic Platforms: Making Systems of Action Work at Enterprise Scale

About the Author :
Jeff Needham is the Field CTO for Insurance at MongoDB, with nearly 30 years of software delivery experience. As former senior director of architecture at Travelers, he led a large-scale modernization of mission-critical systems. Earlier, he held senior roles at Aetna/CVS and other software companies. Today, he advises insurance executives on agentic AI, core systems modernization, and the enterprise data architecture that enables both. He holds a master's in political strategy from George Washington University. Humza Akhtar is Field CTO for Manufacturing and helps global manufacturers and automakers achieve their digital transformation vision with MongoDB. Before MongoDB, he worked at Ernst & Young Canada in the digital operations consultancy practice. After completing his education in Singapore, he worked with the Singapore manufacturing industry for many years on Industry 4.0 research and implementation. He has spent his entire career enabling connected factories and connected cars for global manufacturing and automotive clients. He is a published author on Industry 4.0, and his focus is now on enabling Agentic AI use cases in the industrial sector. Peyman Parsi began his career in financial services software engineering at SS&C, building wealth management software for banking. In 2001, he joined the Toronto Stock Exchange (TSX), where he led the development of capital markets solutions. Over 18 years at TSX, Peyman delivered several large-scale transformations and served as chief technology delivery officer. In 2020, he became CTO at Blanc Labs, focusing on banking and digital lending solutions. Peyman is currently the Global Field CTO for the financial services industry at MongoDB and a member of the advisory committee at the CIO Association of Canada. Albert Cortez is a senior industry solutions consultant for insurance at MongoDB, where he advises executives and architects on modernization and data strategy, supports strategic sales through discovery, workshops, and proof-of-value work, and creates thought leadership while helping shape product direction and go-to-market strategy. Before MongoDB, he held solution and technical architect roles at Accenture and regional consulting firms, leading teams on large-scale digital transformation projects for insurance clients and delivering Salesforce industry-based solutions across Latin America, Europe, and the US. Taylor Hedgecock is a strategic program leader and transformation partner with experience spanning startups to multinationals. At MongoDB, she has led high-impact programs across AI, partner ecosystems, and services modernization, guiding C-level priorities, go-to-market readiness, and large-scale change. She currently serves as senior program manager on the industry solutions team, partnering with ISVs and AI innovators to bring next-generation solutions to market. Previously, as chief of staff for professional services leadership, she helped launch new offerings and guided modernization strategy, shaping MongoDB's vision for applying AI to its hardest problems. Sebastian Rojas Arbulu is a senior industry solutions specialist at MongoDB, collaborating across industries to help customers realize MongoDB's value through tailored, data-driven solutions, particularly for AI integration. He also leads his team's content strategy across blogs, white papers, magazines, and other thought leadership. With a background in IT consulting, marketing, and digital transformation, he has extensive experience identifying customer needs and developing innovative solutions that prepare data for intelligent applications. He holds a Bachelor of Business Administration degree.