Development Setup¶

This guide provides detailed instructions for setting up a development environment for Nevron.

Prerequisites¶

Python 3.13+ (required for the latest features and optimizations)
Poetry (for dependency management)
Make (for using Makefile commands)
Git (for version control)
Docker (optional, for containerized development)

Initial Setup¶

1. Clone the Repository¶

git clone https://github.com/axioma-ai-labs/nevron.git
cd nevron

2. Install Poetry¶

If you haven't installed Poetry yet:

curl -sSL https://install.python-poetry.org | python3 -

Make sure to add Poetry to your PATH according to the installation instructions.

3. Install Dependencies¶

make deps

This will: - Install all project dependencies using Poetry - Set up pre-commit hooks for code quality - Configure the development environment

Environment Configuration¶

Basic Configuration¶

Create a development environment file:

cp .env.dev .env

LLM Provider Configuration¶

You have several options for configuring the LLM provider:

Option 1: Use a Third-Party LLM API¶

Edit your .env file to include one of these LLM providers:

# OpenAI
OPENAI_API_KEY=your_key_here
LLM_PROVIDER=openai

# Anthropic
ANTHROPIC_API_KEY=your_key_here
LLM_PROVIDER=anthropic

# xAI (Grok)
XAI_API_KEY=your_key_here
LLM_PROVIDER=xai

# DeepSeek
DEEPSEEK_API_KEY=your_key_here
LLM_PROVIDER=deepseek

# Qwen
QWEN_API_KEY=your_key_here
LLM_PROVIDER=qwen

# Venice
VENICE_API_KEY=your_key_here
LLM_PROVIDER=venice

Option 2: Use Llama via API Services¶

# Standard Llama API
LLAMA_PROVIDER=llama-api
LLAMA_API_KEY=your_key_here
LLM_PROVIDER=llama

# OpenRouter
LLAMA_PROVIDER=openrouter
LLAMA_API_KEY=your_key_here
LLM_PROVIDER=llama

# Fireworks
LLAMA_PROVIDER=fireworks
LLAMA_API_KEY=your_key_here
LLM_PROVIDER=llama

Option 3: Run Ollama Locally (Recommended for Development)¶

Install Ollama following the instructions at ollama.ai
Pull a small model:
```
ollama pull llama3:8b-instruct
```

Configure your .env file:

LLM_PROVIDER=llama
LLAMA_PROVIDER=ollama
LLAMA_OLLAMA_MODEL=llama3:8b-instruct

Memory Configuration¶

For development, you can use either ChromaDB (default) or Qdrant:

# ChromaDB (lightweight, file-based)
MEMORY_BACKEND_TYPE=chroma

# Qdrant (requires running Qdrant server)
MEMORY_BACKEND_TYPE=qdrant
MEMORY_HOST=localhost
MEMORY_PORT=6333

Running the Application¶

Start the Development Server¶

make run

Run Tests¶

make test

Lint Code¶

make lint

Format Code¶

make format

IDE Setup¶

VSCode¶

Recommended extensions: - Python - Ruff - isort - GitLens - Poetry

Recommended settings (settings.json):

{
    "python.formatting.provider": "ruff",
    "editor.formatOnSave": true,
    "python.linting.enabled": true,
    "python.linting.mypyEnabled": true
}

PyCharm¶

Enable Ruff formatter
Set Python interpreter to the Poetry environment
Enable mypy for type checking
Configure Poetry as the package manager

Troubleshooting¶

Common development issues:

Python Version: Ensure you're using Python 3.13+
```
python --version
```
Poetry Environment: Verify Poetry has created the virtual environment
```
poetry env info
```
Dependencies: Check installed packages
```
poetry show
```
Ollama: If using Ollama, verify it's running
```
ollama list
```
Environment Variables: Ensure your .env file contains the necessary configuration
```
cat .env | grep -v "^#" | grep .
```
Logs: Check the application logs for detailed error messages
```
tail -f logs/app.log
```

For more help, visit our GitHub Discussions or open an issue on the repository.