LeetCodeSolver Crew

Welcome to the LeetCodeSolver project, powered by crewAI! 🚀
This innovative agent-based system is designed to help you solve LeetCode problems effortlessly, from easy to hard. A team of AI agents collaborates to analyze problems, design algorithms, write and test code, and optimize performance.

How the Crew Works

The LeetCodeSolver Crew operates in a hierarchical structure to efficiently solve LeetCode problems. Here's a breakdown of the workflow:

1. Manager: The manager receives the LeetCode problem and delegates tasks to the appropriate team members.
2. Software Architect: Analyzes the problem and designs the algorithm.
3. Software Engineer: Writes the Python code based on the algorithm.
4. Tester: Writes test cases to ensure the code works correctly.
5. QA: Reviews the code and test cases, providing reports and suggestions for improvements.

Each team member focuses on their specialized role, ensuring a streamlined and efficient problem-solving process.

---

🚀 Getting Started

Prerequisites

Ensure you have Python >=3.10 <3.13 installed. This project uses UV for seamless dependency management.

Installation Steps

1. Install Docker

• Docker is necessary to execute Python code using the CodeInterpreterTool.

2. Install UV
   If you haven't already, install UV by running:

   pip install uv

3. Set Up the Project
   Navigate to your project directory and install dependencies:

   uv install

   (Optional) Lock dependencies and install them using:

   crewai install

---

⚙️ Configuration

1. Environment Setup

   • Copy .env-template to .env.
   • Configure the LLM_PROVIDER and MODEL variables by commenting/uncommenting the options.
   • Add your API key for the chosen LLM provider (e.g., OPENAI_API_KEY for OpenAI).

2. Prepare LeetCode Problem Files

   • Visit LeetCode and copy the problem description into a markdown file.
   • Save the markdown file in the data folder.
   • Update the .env file with the path to your problem file:

     LEETCODE_FILE="data/your_leetcode_file.md"

---

🔍 Monitoring Progress with Phoenix

1. Open a new terminal window and start Phoenix:

   phoenix serve

2. Access the Phoenix dashboard at: http://localhost:6006

---

🏃 Running the Project

To start solving problems with your AI crew, run the following command from the project root:

crewai run

• This will initialize the LeetCodeSolver Crew, assign tasks, and begin execution based on your configuration.
• The final code will be stored in result folder with the same name as leetcode file.

---

🤖 Meet Your Crew

The LeetCodeSolver Crew consists of multiple specialized AI agents, each with unique roles and tools. These agents work together to tackle tasks defined in:

• config/tasks.yaml: Task definitions and workflows.
• config/agents.yaml: Agent capabilities and configurations.

---

Unleash the power of AI collaboration and solve LeetCode problems like never before! 🎉