Skip to content

Quickstart Guide

This guide will help you get Nevron, your autonomous AI agent, running in 5 simple steps.

Prerequisites

To be able to run the agent, you need to have the following:

  • Python 3.12
  • Pipenv
  • Docker (optional, for Qdrant memory backend)
  • OpenAI API key (or any other API key of an LLM provider of your choice)

Setup in 5 Steps

1. Clone & Install

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

# install dependencies
make deps

2. Configure Environment

# copy example environment file
cp .env.dev .env

Required environment variables: 

ENVIRONMENT=development         # Set environment (development or production)
OPENAI_API_KEY=your_key_here    # Required for embeddings

Note, there're many more environment variables you can set! Please, explore the .env.dev file.

3. Choose Memory Backend

Option A: Chroma (Default)

No additional setup required. Uses local file storage.

Option B: Qdrant

# create storage directory
mkdir qdrant_storage

# run qdrant container
docker run -p 6333:6333 -p 6334:6334 \
    -v $(pwd)/qdrant_storage:/qdrant/storage:z \
    qdrant/qdrant

Update .env: 

MEMORY_BACKEND_TYPE=qdrant

4. Configure Nevron's Personality

Setup the personality, goals and rest time of your agent depending on your needs.

In .env: 

AGENT_PERSONALITY="A helpful AI assistant focused on research and analysis"
AGENT_GOAL="To assist with information gathering and analysis"
AGENT_REST_TIME=300  # seconds between actions

5. Run Nevron

make run

Available Workflows

Nevron comes with two pre-configured workflows which can be used as a starting point:

  • Analyze signal: Processes and analyzes incoming signal data
  • Research news: Gathers and analyzes news using Perplexity API

If you want to create your own workflows, or want to learn more about how the workflows work, please refer to the Workflows documentation.

Customization

For more customization you can add your custom workflows and tools, adjust planning parameters, switch LLM providers, fine-tune the hyper-parameters, etc.

Please refer to the Agent for more information on how to customize the agent, its behavior & personality.

Troubleshooting

  • Ensure all required API keys are set in .env
  • Check logs in the console for detailed error messages
  • Verify Python version: python --version
  • Confirm dependencies: pipenv graph

If you have any questions or need further assistance, please refer to the GitHub Discussions.