📊 Multi-Agent Financial Research Assistant

A serverless, AI-powered assistant for financial research that fetches live market news and data, performs sentiment analysis, and generates investor-friendly reports. Built with LangGraph, LangChain, Streamlit, AWS Lambda, and DynamoDB.

---

🚀 Features

• Agent A: Fetches live market news & stock data via APIs.
• Agent B: Summarizes text and performs sentiment analysis.
• Agent C: Generates investor-friendly insights and reports using GPT-4.
• LangGraph Orchestration: Manages agent workflows.
• AWS Lambda + DynamoDB: Enables serverless execution & persistence.
• Streamlit Dashboard: User-friendly interface for exploring insights.

---

📂 Project Structure

├── agents
│   ├── agent_a_fetcher.py      # Fetch live financial data & news
│   ├── agent_b_analyzer.py     # Summarization & sentiment analysis
│   └── agent_c_reporter.py     # Report generation
├── orchestration
│   └── langgraph_orchestrator.py   # Orchestrates agent workflows
├── lambda
│   └── lambda_handler.py       # AWS Lambda handler
├── utils
│   └── dynamo.py               # DynamoDB helper functions
├── streamlit_app.py            # Streamlit dashboard
├── run_local.py                # Local runner for development
├── requirements.txt            # Python dependencies
├── .env.example                # Example environment variables
└── README.md                   # Project documentation

---

⚙️ Setup

1. Clone Repository

git clone https://github.com/hq969/multi-agent-financial-assistant.git
cd multi-agent-financial-assistant

2. Install Dependencies

pip install -r requirements.txt

3. Configure Environment Variables

Copy .env.example to .env and update with your credentials:

OPENAI_API_KEY=your-openai-api-key
NEWS_API_KEY=your-newsapi-key
MARKET_API_KEY=your-alpha-vantage-key
DYNAMO_TABLE=FinancialReports

---

▶️ Usage

Run Locally

python run_local.py

Streamlit Dashboard

streamlit run streamlit_app.py

Then open http://localhost:8501 in your browser.

Deploy to AWS Lambda

• Package with dependencies.
• Set environment variables in Lambda.
• Ensure DynamoDB table exists (FinancialReports).
• Deploy handler: lambda/lambda_handler.lambda_handler.

---

📊 Example Workflow

1. Agent A fetches stock news + financial data.
2. Agent B generates a summary + sentiment analysis.
3. Agent C compiles reports for investors.
4. Orchestrator pipelines results to DynamoDB & UI.

---

🛠️ Tech Stack

• Frontend: Streamlit
• Backend: AWS Lambda (Python)
• Data Storage: DynamoDB
• AI/LLM: OpenAI GPT-4 via LangChain
• Workflow: LangGraph
• APIs: NewsAPI, Alpha Vantage

---

✅ Next Steps

• Add unit tests in a tests/ folder.
• Create CI/CD pipeline for Lambda + Streamlit.
• Add Mermaid architecture diagram to README.

---

👨‍💻 Author

Built by Harsh Sonkar ⚡

---