Quick Start

This guide will get you up and running with the Moderately AI Python SDK in minutes.

Setup

First, install the SDK:

pip install moderatelyai-sdk

Get your API credentials from the Moderately AI platform:

API Key: Your authentication token
Team ID: Your team identifier

Environment Variables

Set your credentials as environment variables (recommended):

export MODERATELY_API_KEY="your-api-key-here"
export MODERATELY_TEAM_ID="your-team-id-here"

Or create a .env file:

MODERATELY_API_KEY=your-api-key-here
MODERATELY_TEAM_ID=your-team-id-here

Basic Usage

Synchronous Client

import moderatelyai_sdk

# Initialize client with environment variables
client = moderatelyai_sdk.ModeratelyAI()

# Or initialize with explicit credentials
client = moderatelyai_sdk.ModeratelyAI(
    team_id="your-team-id",
    api_key="your-api-key"
)

# List users in your team
users = client.users.list()
print(f"Found {len(users)} users")

# Create a dataset
dataset = client.datasets.create(
    name="My First Dataset",
    description="A test dataset"
)

# Upload a file
file = client.files.upload(
    file="path/to/your/file.csv",
    name="Training Data"
)

Asynchronous Client

import asyncio
import moderatelyai_sdk

async def main():
    # Use async context manager (recommended)
    async with moderatelyai_sdk.AsyncModeratelyAI() as client:
        # Same operations, just with await
        users = await client.users.list()
        print(f"Found {len(users)} users")

        dataset = await client.datasets.create(
            name="My Async Dataset",
            description="A test dataset created asynchronously"
        )

        file = await client.files.upload(
            file="path/to/your/file.csv",
            name="Async Training Data"
        )

# Run the async function
asyncio.run(main())

Working with Files

Files are one of the most common resources you’ll work with:

# Upload a file
file = client.files.upload(
    file="/path/to/document.pdf",
    name="Important Document",
    metadata={"category": "legal", "priority": "high"}
)

# Check file status
if file.is_ready():
    print(f"File ready: {file.name} ({file.file_size} bytes)")

    # Download file content
    content = file.download()  # To memory
    file.download(path="./local_copy.pdf")  # To disk

    # Check file type
    if file.is_document():
        print("This is a document file")
    elif file.is_csv():
        print("This is a CSV file")

# Clean up
file.delete()

Error Handling

Always handle errors appropriately:

from moderatelyai_sdk import (
    ModeratelyAI,
    APIError,
    AuthenticationError,
    NotFoundError
)

client = ModeratelyAI()

try:
    dataset = client.datasets.create(name="Test Dataset")
    print(f"Created dataset: {dataset['datasetId']}")
except AuthenticationError:
    print("Invalid API key or insufficient permissions")
except APIError as e:
    print(f"API error occurred: {e}")
    if hasattr(e, 'status_code'):
        print(f"Status code: {e.status_code}")

Next Steps

Learn about File Operations for detailed file handling
Explore datasets for data management
Check out agents for AI agent integration
See async_usage for async patterns
Review the complete Client Classes API reference

For complete examples, visit the examples directory in the repository.