Complete User Guide

Step-by-step guide to using the AI Hedge Fund application. Learn everything you need to know to build trading flows, configure AI agents, and analyze stocks.

🚀 Open Application View Documentation

🚀 Quick Start

First Time? Follow these steps to get started in 5 minutes.
1

Access the Application

Open the AI Hedge Fund application in your browser. The application runs at http://localhost:5173 by default.

💡 Tip: If you haven't started the application yet, follow the installation guide first.
2

Configure API Keys

Before you can use AI agents, you need to configure at least one LLM API key:

  1. Click on Settings in the top menu
  2. Go to API Keys section
  3. Click "Add API Key"
  4. Select your provider (OpenAI, Groq, Anthropic, or Ollama for free local models)
  5. Enter your API key and save
Recommended: Start with Ollama (free) or GPT-4o-mini (very affordable) for testing.
3

Build Your First Trading Flow

Create a simple trading flow to analyze stocks:

  1. Drag a "Tickers" node onto the canvas
  2. Configure it with stock symbols (e.g., AAPL, MSFT)
  3. Drag an AI agent (e.g., Warren Buffett Agent) onto the canvas
  4. Connect the Tickers node to the Agent node
  5. Add a Risk Manager node
  6. Add a Portfolio Manager node
  7. Connect: Tickers → Agent → Risk Manager → Portfolio Manager
4

Configure and Run

Set up your flow and run the analysis:

  1. Click on the Portfolio Manager node
  2. Set initial cash (e.g., $100,000)
  3. Set margin requirement (e.g., 50%)
  4. Click the "Run" button
  5. Watch the analysis progress in real-time
  6. View results in the bottom panel

🖥️ Understanding the Interface

The AI Hedge Fund interface consists of several key areas:

📋 Left Sidebar

  • Agent Library: All 18 AI agents available to drag onto canvas
  • Flow Management: Save, load, and manage your trading flows
  • Templates: Pre-built flow templates to get started quickly

🎨 Center Canvas

  • Visual Flow Builder: Drag-and-drop interface
  • Node Connections: Connect nodes with edges (arrows)
  • Zoom & Pan: Navigate large flows easily

⚙️ Right Panel

  • Node Configuration: Configure selected nodes
  • Tickers: Set which stocks to analyze
  • Model Selection: Choose LLM model for agents
  • Parameters: Agent-specific settings

📊 Bottom Panel

  • Execution Results: View analysis results
  • Logs: Real-time progress updates
  • Trading Decisions: See buy/sell/hold signals
  • Portfolio Status: Current positions and cash

🔧 Building Trading Flows

Basic Flow Structure

Every trading flow follows this pattern:

Start
  ├─ Tickers Node (AAPL, MSFT, etc.)
  ├─ Analyst Agents (parallel)
  │   ├─ Warren Buffett Agent
  │   ├─ Peter Lynch Agent
  │   └─ Technicals Agent
  │       ↓
  Risk Manager (aggregates signals)
      ↓
  Portfolio Manager (final decisions)
      ↓
End

Step-by-Step: Building a Flow

Step 1: Add Tickers Node

  1. Drag "Tickers" from the left sidebar onto the canvas
  2. Click on the node to select it
  3. In the right panel, enter stock symbols: AAPL, MSFT, NVDA
  4. Separate multiple tickers with commas
Free Tickers: AAPL, GOOGL, MSFT, NVDA, TSLA don't require API keys. Others may need Financial Datasets API.

Step 2: Add Analyst Agents

  1. Drag AI agents from the sidebar (e.g., Warren Buffett, Peter Lynch)
  2. Place them on the canvas
  3. Connect the Tickers node to each agent (drag from Tickers output to agent input)
  4. Configure each agent:
    • Select LLM model
    • Set any agent-specific parameters
💡 Tip: Start with 2-3 agents. More agents = more analysis but higher API costs.

Step 3: Add Risk Manager

  1. Drag "Risk Manager" onto the canvas
  2. Connect all analyst agents to the Risk Manager
  3. Configure risk parameters:
    • Maximum position size
    • Risk limits
    • Portfolio constraints
⚠️ Important: Always include a Risk Manager in your flows to protect your portfolio.

Step 4: Add Portfolio Manager

  1. Drag "Portfolio Manager" onto the canvas
  2. Connect Risk Manager to Portfolio Manager
  3. Configure portfolio settings:
    • Initial Cash: Starting capital (e.g., $100,000)
    • Margin Requirement: Percentage (e.g., 50%)
    • Starting Positions: Optional existing positions

Step 5: Run the Flow

  1. Click the "Run" button in the top bar
  2. Watch real-time progress in the bottom panel
  3. See each agent analyze stocks
  4. View trading signals and reasoning
  5. Review final portfolio decisions

🚀 Advanced Flow Patterns

Multi-Strategy Analysis

Run multiple agents in parallel for comprehensive analysis:

Multiple Analyst Agents (parallel)
    ↓
Risk Manager (aggregates signals)
    ↓
Portfolio Manager

Use Case: Get diverse perspectives on the same stocks.

Sequential Analysis

Chain agents where one feeds into the next:

Fundamentals Agent
    ↓
Valuation Agent (uses fundamentals)
    ↓
Risk Manager
    ↓
Portfolio Manager

Use Case: Build complex analysis pipelines.

Common Strategy Patterns

Conservative

Warren Buffett + Ben Graham + Risk Manager

Focus on value with strong risk management

Growth

Peter Lynch + Cathie Wood + Risk Manager

High-growth companies with innovation

Balanced

Multiple value + growth agents + Risk Manager

Diversified analysis approach

📊 Backtesting Strategies

Test your strategies on historical data to see how they would have performed.

Setting Up a Backtest

  1. Build your trading flow (same as regular analysis)
  2. Click the "Backtest" button
  3. Set date range:
    • Start Date: e.g., 2024-01-01
    • End Date: e.g., 2024-12-31
  4. Set initial capital (e.g., $100,000)
  5. Click "Run Backtest"

Understanding Backtest Results

Performance Metrics

  • Total Return: Percentage gain/loss
  • Sharpe Ratio: Risk-adjusted returns
  • Max Drawdown: Largest peak-to-trough decline
  • Win Rate: Percentage of profitable trades

Portfolio Details

  • Daily Portfolio Value: Chart over time
  • Final Positions: Stocks held at end
  • Final Cash: Remaining cash
  • Trade History: All trades made
💡 Best Practices:
  • Test on different time periods
  • Compare multiple strategies
  • Consider market conditions
  • Review individual trade decisions

💾 Saving and Loading Flows

Saving a Flow

  1. Build your trading flow
  2. Click the "Save" button in the top bar
  3. Enter a name for your flow
  4. Optionally mark as a template
  5. Click "Save"

Loading a Flow

  1. Click the "Load" button
  2. Select a flow from the list
  3. The flow loads onto the canvas
  4. Modify as needed
💡 Tip: Save common flow patterns as templates for quick reuse.

🔧 Troubleshooting

"API Key Not Found"

Solution: Go to Settings → API Keys and add at least one API key. Make sure the key is correct and has credits.

"Flow Won't Run"

Check:

  • All nodes are properly connected
  • Tickers are set correctly
  • API keys are configured
  • Portfolio Manager has initial cash set

"Agents Not Analyzing"

Solution: Check that agents are connected to the Tickers node and have valid API keys configured.

Ready to Start Trading?

Now that you understand how to use AI Hedge Fund, access the application and start building your trading strategies!

🚀 Access Application Now
Explore All Agents Full Documentation FAQ