In Episode 111 of Azure Cosmos DB TV, host Mark Brown is joined by Yohan Lasorsa to explore how developers can build agent-powered applications using a fully serverless architecture. This episode focuses on a practical, end-to-end example that demonstrates how transactional application data and AI-driven experiences can coexist on a single platform without introducing additional infrastructure or operational overhead.

The session walks through a sample application inspired by a familiar business scenario: a simple ordering system. What begins as a traditional REST-based business API evolves into an agent-enabled experience, where an AI agent can reason over real application data, maintain conversational memory, and take meaningful actions on behalf of users.

A Practical Agentic Architecture

At the core of the sample is Azure Cosmos DB, which serves as both the system of record for transactional data and the backbone for AI-powered interactions. Orders and business entities are stored in Azure Cosmos DB, enabling low-latency reads and writes while providing a consistent data model that can be reused across application layers.

The episode introduces an MCP (Model Context Protocol) bridge that exposes business data to AI agents in a structured and secure way. Rather than hard-coding database access into the agent itself, the MCP layer acts as an abstraction that allows the agent to discover available data and operations dynamically. This makes it easier to evolve the application over time while keeping the agent logic clean and portable.

LangChain.js is used to implement the agent, providing orchestration, reasoning, and chat memory. By combining LangChain.js with the MCP bridge, the agent can answer questions, retrieve order details, and reason about business context using the same data that powers the transactional API.

Walking Through the Sample

During the live demo, Mark and Yohan walk through the application architecture and then dive into the codebase. They show how the REST API is structured, how Azure Cosmos DB is accessed from the service layer, and how the MCP server exposes those capabilities to the agent. The discussion highlights how little additional code is required to move from a traditional API to an agent-enabled experience.

A key takeaway from the demo is how chat memory is handled. Rather than introducing a separate system for conversational state, the sample uses existing platform components to persist context in a scalable and reliable way. This reinforces the idea that agentic applications do not need an entirely new stack—they can be built by extending patterns developers already know.

Extending the Use Case

The episode closes with a discussion of possible extensions to the sample. These include adding richer queries, expanding agent capabilities, and taking advantage of additional Azure Cosmos DB features as the application grows. The same foundation can support more advanced scenarios, such as multi-agent workflows, deeper analytics, or integration with other AI services.

By keeping everything on a serverless footprint, the architecture remains easy to deploy, operate, and scale. Developers can move quickly from prototype to production without rethinking their data layer or introducing unnecessary complexity.

Watch the Episode

If you’re interested in building agentic applications that combine reliable transactional data with AI-driven interactions, this episode provides a clear and approachable blueprint.

Watch the full episode here: https://youtu.be/7faP1YPOFCA

Explore the related resources and sample code: https://aka.ms/lcjs/agent-mcp

Leave a Review

Tell us about your Azure Cosmos DB experience! Leave a review on PeerSpot and get a $50 gift. Get started here.

About Azure Cosmos DB

Azure Cosmos DB is a fully managed and serverless NoSQL and vector database for modern app development, including AI applications. With its SLA-backed speed and availability as well as instant dynamic scalability, it is ideal for real-time NoSQL and MongoDB applications that require high performance and distributed computing over massive volumes of NoSQL and vector data.

To stay in the loop on Azure Cosmos DB updates, follow us on X, YouTube, and LinkedIn.