Skip to content
 
 

Repository files navigation

Ridges AI - SN62

Ridges develops open source software engineering agents.

For Miners

As a miner, your

For Validators

Prerequisites

Local Subtensor Network Setup

Before proceeding with the Ridges AI setup, ensure you have a local Subtensor network running on your machine. Follow the official setup instructions:

Critical Requirements:

  • Create separate wallets for miner and validator operations
  • Ensure both wallets are properly configured and secured

Network Registration and Funding

Complete the following steps to register and fund your wallets on the local subnet:

# Fund validator wallet
btcli wallet faucet --wallet.name validator --subtensor.chain_endpoint ws://127.0.0.1:9945

# Register validator on subnet
btcli subnet register --wallet.name validator --subtensor.chain_endpoint ws://127.0.0.1:9945

# Add stake to validator (required for validation operations)
btcli stake add --wallet.name validator --wallet.hotkey default --subtensor.chain_endpoint ws://127.0.0.1:9945

# Fund miner wallet
btcli wallet faucet --wallet.name miner --wallet.hotkey default --subtensor.chain_endpoint ws://127.0.0.1:9945

# Register miner on subnet
btcli subnet register --wallet.name miner --wallet.hotkey default --subtensor.chain_endpoint ws://127.0.0.1:9945

Development Environment Setup

Repository Initialization

# Clone the repository
git clone https://github.com/ridgesai/ridges.git
cd ridges

# Initialize Python virtual environment
uv venv --python 3.11
source .venv/bin/activate

# Install project dependencies
uv pip install -e .

Local Network Configuration

If running against a local Subtensor instance, set the following environment variable:

export SUBTENSOR_ADDRESS=ws://127.0.0.1:9945

Platform Infrastructure Setup

AWS Configuration

The platform requires the following AWS resources:

  1. AWS Account & CLI Authentication

    • Valid AWS account with appropriate permissions
    • AWS CLI configured and authenticated (with aws configure)
  2. Database Setup

    • Create PostgreSQL RDS instance
    • Apply schema from api/src/db/postgres_schema.sql
  3. Storage Setup

    • Create S3 bucket for file storage
  4. API Integration

Platform Deployment

# Navigate to API directory
cd api

# Configure environment variables
cp api/.env.example api/.env
# Edit api/.env with your configuration values

# Start the platform services
uvicorn api.src.main:app

Validator Environment Setup

Docker Configuration

Build sandbox execution environment and proxy

docker build -t sandbox-runner validator/sandbox
docker build -t sandbox-nginx-proxy validator/sandbox/proxy/

Running the validator

uv run validator/main.py

Miner Agent Development

Agent Upload Process

  1. Access the platform API documentation: http://localhost:8000/docs
  2. Use the /upload/agent endpoint to deploy your agent

Agent Requirements

Your agent file must meet the following criteria for successful deployment:

  • File Structure: Single Python file named agent.py
  • Code Quality: Valid Python syntax and logic
  • Entry Point: Contains an agent_main function at the top level
  • Dependencies: Only imports from:
    • Python standard library
    • Approved libraries specified in api/src/utils/config.py

Note: Files that do not meet these requirements will be rejected during the upload process.

About

Building Software Agents On Bittensor

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages