AI Stock Analyst

AI Stock Analyst is a modern, multi-agent web application that leverages AI to deliver, real-time stock analysis and investment recommendations. The platform integrates specialized AI agents

Comment

🚀 Inspiration

We were inspired by the complexity and fragmentation that everyday investors face when analyzing stocks across global markets. Navigating financial platforms, interpreting trends, and filtering through overwhelming news can be daunting—especially for those without a financial background.

We wanted to reimagine the stock analysis experience by building a smart, agent-powered assistant that acts like a team of virtual analysts, bringing clarity, personalization, and global market access into one intuitive dashboard.

⸻

💼 What it does

AI Stock Analyst: Global Market Companion enables users to analyze stocks from major global markets—USA, India, UK, Germany, and Japan—by simply selecting a ticker and region.

Once initiated, our system: • 📊 Fetches real-time market data (price, volume, currency, and trends) • 📈 Performs technical analysis (RSI, volatility, moving averages) • 📰 Pulls and summarizes recent news relevant to the stock • 🤖 Provides AI-driven investment insights using sentiment and data trends • 📁 Offers portfolio allocation suggestions based on the user’s risk profile

If a stock is entered under the wrong region, our system auto-detects and guides the user toward the correct region to ensure accurate data.

⸻

🏗️ How we built it

We built a multi-agent system powered by Python and Streamlit, orchestrated by a core StockOrchestrator that manages five specialized agents: • NewsAgent (NewsAPI): Retrieves and formats real-time news based on ticker and region • TrendsAgent: Analyzes technical indicators like RSI, price trends, and volatility • DataAgent (Yahoo Finance): Fetches real-time stock data and performs region fallback • InsightsAgent: Uses LLMs to provide AI-generated investment commentary • PortfolioAdvisor: Creates allocation plans based on user risk profiles

We designed the UI using Streamlit with an emphasis on clarity and responsiveness. The agents communicate through structured prompts and lightweight APIs. We deployed the app on Google Cloud Run with CI/CD using Cloud Build.

⸻

🧗 Challenges we ran into • Mapping ticker symbols to correct global exchanges and suffixes • Ensuring graceful fallbacks when data was missing or delayed • Handling noisy or malformed text data from news APIs • Structuring agent interactions to avoid hallucinations or redundant analysis • Keeping the UI clean while presenting diverse data types side-by-side

⸻

🏅 Accomplishments that we’re proud of • Built a globally aware stock analyzer with intelligent region detection • Designed a modular agent system with real-time decision-making • Delivered clean, readable AI investment guidance from structured and unstructured data • Streamlined the end-to-end deployment pipeline using GCP tooling • Created a UI that’s both data-rich and beginner-friendly

⸻

📚 What we learned • The importance of validating financial data across multiple regions • How to apply LLMs responsibly for investment advisory use cases • Techniques to summarize news articles for relevance and readability • Real-world considerations for deploying Streamlit apps with CI/CD in production • Strategies for designing cooperative multi-agent systems with specialized roles

⸻

🔮 What’s next for AI Stock Analyst: Global Market Companion

We’re excited to expand the project in the following directions: • 🌍 Support for additional global exchanges (e.g., Canada, Australia, Hong Kong) • 📆 Historical performance comparison across multiple tickers • 🗂️ Exportable insights in PDF format or via email summaries • 📊 Sentiment charts from news and social media • 📦 Integrating user authentication for personalized dashboards • 📱 Optimized mobile UI and voice-based interactions (via AI assistant SDKs)

Built With

Share this project:

Updates