🤖 Agentic AI for Engineers - Build your own Financial Advisor with MCP

Retail investors and wealth management platforms are increasingly exploring AI-driven advisory tools to support portfolio construction, market analysis, and long-term investment decisions.
But building a reliable financial AI advisor requires far more than connecting an LLM to stock prices.

How do you design an Agentic AI system that can:

• Retrieve and structure financial market data,
• Reason over portfolio allocation and risk,
• Maintain state and context across interactions,
• And provide structured, explainable investment insights?

In this 60-minute interactive webinar, we’ll walk through the end-to-end design and implementation of an Agentic AI Financial Advisor, combining structured financial data with reasoning agents powered by LLMs.

The session includes a live implementation using Python, LangChain, OpenAI models, MCP protocol, and Yahoo Finance data, demonstrating how to orchestrate tools, memory, and reasoning steps into a coherent financial advisory system.

***

### You’ll learn

Core Concepts of Agentic AI in Finance
How agent-based systems differ from simple chatbots:

• Tool use and financial data retrieval
• Multi-step reasoning over market signals
• State management across sessions
• Guardrails and responsible AI considerations

Using MCP Protocol for Structured Agent Orchestration
How MCP enables:

• Modular agent communication
• Tool abstraction and extensibility
• Separation of reasoning and execution layers
• Scalable financial agent architectures

Integrating Real Market Data (Yahoo Finance)
How to:

• Retrieve live and historical market data
• Structure time-series inputs for analysis
• Perform portfolio-level reasoning
• Generate explainable investment summaries

Live Demo: Building a Financial Advisor Agent
Step-by-step implementation:

• Setting up LangChain with OpenAI
• Connecting to Yahoo Finance APIs
• Implementing MCP-based tool orchestration
• Generating portfolio analysis and allocation suggestions
• Producing transparent, plain-text financial explanations

From Data to Advice: Responsible AI & Limitations
Discussion around:

• Risk, uncertainty, and hallucination management
• Compliance considerations
• Human-in-the-loop investment workflows
• Where Agentic AI adds value in wealth management

***

Duration: 60 minutes
Level: Intermediate (basic Python knowledge recommended)
Tech Stack: Python, LangChain, OpenAI, MCP protocol, Yahoo Finance
URL: https://events.teams.microsoft.com/event/0dc10ae2-1ad4-4435-9239-6fd1db7be98a@d94ea0cb-fd25-43ad-bf69-8d9e42e4d175