Building an Intelligent Stock Assistant with pydantic-ai and Groq LLM

Project Overview

Developed a financial assistant chatbot that combines LLM capabilities with real-time stock market data. The system processes natural language queries to provide current stock prices and market insights.

Key Features

Natural language processing using Groq LLM
Real-time stock data integration via yfinance
Type-safe data handling with Pydantic models
Interactive web interface built with Gradio
Custom agent configuration with tool integration

Technical Implementation

The project leverages several key technologies to create a robust and user-friendly stock assistant:

Data Model and Type Safety

from pydantic import BaseModel

class StockPriceResult(BaseModel):
    symbol: str
    price: float
    currency: str = "USD"
    message: str

LLM Agent Configuration

stock_agent = Agent(
    "groq:llama3-groq-70b-8192-tool-use-preview",
    result_type=StockPriceResult,
    system_prompt="You are a helpful financial assistant specialized in providing stock market information and insights."
)

Stock Data Integration

import yfinance as yf

def get_stock_price(symbol: str) -> StockPriceResult:
    ticker = yf.Ticker(symbol)
    data = ticker.info
    
    return StockPriceResult(
        symbol=symbol,
        price=data['regularMarketPrice'],
        message=f"Current price for {symbol} is ${data['regularMarketPrice']:.2f}"
    )

Gradio Interface

import gradio as gr

def process_query(user_input: str) -> str:
    result = stock_agent.run(user_input)
    return result.message

interface = gr.Interface(
    fn=process_query,
    inputs=gr.Textbox(label="Ask about any stock"),
    outputs=gr.Textbox(label="Response"),
    title="Stock Market Assistant",
    description="Get real-time stock information and insights"
)

Key Achievements

Implemented type-safe data handling using Pydantic models
Integrated real-time stock market data through yfinance API
Created an intuitive chat interface using Gradio
Leveraged Groq’s powerful LLM for natural language understanding
Developed a modular and extensible architecture

Future Enhancements

Add historical price analysis capabilities
Implement portfolio tracking features
Include market sentiment analysis
Expand to multiple financial markets
Add technical indicator calculations

The Stock Assistant demonstrates the power of combining modern LLM capabilities with real-time financial data, creating a practical tool for both casual investors and financial professionals.

Technologies