Quick Start

Viewing typescript

switch to python

Prerequisites

Node.js: version 20+ (opens in a new tab)

OS: macOS or Linux (WSL supported for Windows)
Run Docker: Run open -a Docker to start Docker Desktop. Installation instructions (opens in a new tab)

Create Your Moose Project

Initialize a New Project

Terminal

npx create-moose-app@latest my-moose-app ts

Terminal

# Create and activate a virtual environment (recommended)
python3 -m venv .venv && source .venv/bin/activate
 
# Install CLI and initialize project
pip install -U moose-cli && moose-cli init my-moose-app python

This creates a new project with all necessary files and configuration.

Set Up Project Dependencies

Terminal

cd my-moose-app && npm install

Start the Development Server

Terminal

npx moose-cli dev

Verify Infrastructure Setup

Open a new terminal, navigate to your project directory, and run:

Terminal

npx moose-cli ls

You should see this output showing your ingestion points and tables:

Data Model	Ingestion Point	Table
Foo	/ingest/Foo	Foo_0_0
Bar	/ingest/Bar	Bar_0_0

Understanding Data Models

Explore Data Models

Your project includes two sample data models in app/datamodels/models.ts:

/app/datamodels/models.ts

import { Key } from "@514labs/moose-lib";
 
export interface Foo {
  primaryKey: Key<string>;
  timestamp: Date;
  optionalString?: string;
}
 
export interface Bar {
  primaryKey: Key<string>;
  utcTimestamp: Date;
  hasString: boolean;
  stringLength: number;
}

Your project includes two sample data models in app/datamodels/models.py:

/app/datamodels/models.py

from moose_lib import Key, moose_data_model
from dataclasses import dataclass
from datetime import datetime
from typing import Optional
 
@moose_data_model
@dataclass
class Foo:
  primary_key: Key[str]
  timestamp: datetime
  optional_string: Optional[str]
 
@moose_data_model
@dataclass
class Bar:
  primary_key: Key[str]
  utc_timestamp: datetime
  has_string: bool
  string_length: int

MooseTip:

Moose automatically creates database tables and ingestion endpoints for these models. Run moose-cli lsnpx moose-cli ls to see them.

Ingest Data

Send Data to an Ingestion Endpoint

Terminal

curl -X POST "http://localhost:4000/ingest/Foo" \
  -H "Content-Type: application/json" \
  -d '{"primaryKey": "1234567890", "timestamp": 1546300801.0, "optionalString": "hello world"}'

Verify Data Ingestion

Terminal

npx moose-cli peek Foo

You should see your ingested data:

{"primaryKey": "1234567890", "timestamp": 1546300801.0, "optionalString": "hello world"}

Troubleshooting

Next Steps

Congratulations!

You've successfully set up a Moose project and ingested your first data.

Create Ingestion Workflows

Build data pipelines with Moose Workflows

Stream Processing

Process data streams with Streaming Functions

Create Analytics APIs

Build and expose fully-typed APIs to consume your processed data

Database Connectivity

Connect to the local ClickHouse database:

moose.config.toml

[clickhouse_config]
db_name = "local"
user = "panda"
password = "pandapass"
host = "localhost"
host_port = 18123
native_port = 9000
use_ssl = false

VSCode Integration

Install SQLTools (opens in a new tab) and SQLTools ClickHouse Driver (opens in a new tab)
Moose automatically configures these extensions for your project

Introduction Project Structure