Devika: open-source AI software engineer intelligence that understands, splits instructions into subtasks and writes code

General Introduction

Devika is an advanced AI software engineer that understands high-level human instructions, breaks them down into steps, researches relevant information, and writes code to achieve a given goal. It intelligently develops software using large-scale language models, planning and reasoning algorithms, and web browsing capabilities.Devika aims to be an open-source alternative to Cognition AI's Devin, providing an AI programming assistant capable of completing complex coding tasks with minimal human guidance.

Function List

• Support for Claude 3, GPT-4, Gemini, Mistral, Groq and native LLMs
• Advanced AI planning and reasoning skills
• Keyword extraction for focused research
• Seamless web browsing and information gathering
• Code writing in multiple programming languages
• Dynamic agent state tracking and visualization
• Natural language interaction through the chat interface
• Project-based organization and management
• Extensible architecture to add new features and integrations

Using Help

Installation process

1. Cloning Devika Warehouse:

   git clone https://github.com/stitionai/devika.git

2. Go to the project catalog:

   cd devika
   

3. Create a virtual environment and install the required dependencies:

   uv venv
   source .venv/bin/activate # macOS and Linux
   .venv\Scripts\activate # Windows
   uv pip install -r requirements.txt
   

4. Install Playwright to enable browser functionality:

   playwright install --with-deps
   

5. Start the Devika server:

   python devika.py
   

   If all is well, you will see the following output:

   root: INFO : Devika is up and running!
   

6. Start the front end:

   cd ui/
   bun install
   bun run start
   

7. Open your browser and navigate tohttp://127.0.0.1:3001Visit the Devika web interface.

Guidelines for use

1. Open the Devika web interface.
2. To create a new project, click "Select Project" and then "New Project".
3. Select the search engine and model configuration.
4. Provide a high-level goal or mission description in the chat screen.
5. Devika will process your request, break it down into steps, and start performing tasks.
6. Monitor Devika's progress, view generated code, and provide additional guidance or feedback as needed.
7. Once Devika completes the task, review the generated code and project files.
8. Iterate and refine the project by providing further instructions or modifications.