simple-examples

Hatchet Simple Examples

This is an example project demonstrating how to use Hatchet with Python.

Prerequisites

Before running this project, make sure you have the following:

1. Python 3.10 or higher installed on your machine.
2. Poetry package manager installed. You can install it by running pip install poetry, or by following instructions in the Poetry Docs

Setup

1. Create a .env file in the ./backend directory and set the required environment variables. This requires the HATCHET_CLIENT_TOKEN variable created in the Getting Started README. If you would like to try the Generative AI examples in ./src/genai You will also need, a OPENAI_API_KEY which can be created on the OpenAI Website.

If you're running Hatchet locally without TLS:

HATCHET_CLIENT_TLS_STRATEGY=none
HATCHET_CLIENT_TOKEN="<token>"
OPENAI_API_KEY="<openai-key>" # (OPTIONAL) only required to run examples in [./src/genai](./src/genai)

If you're using Hatchet Cloud:

HATCHET_CLIENT_TOKEN="<token>"
OPENAI_API_KEY="<openai-key>" # (OPTIONAL) only required to run examples in [./src/genai](./src/genai)

2. Open a terminal and navigate to the project root directory (/simple-examples).

3. Run the following command to install the project dependencies:

poetry install

Running the Hatchet Worker

Next, start the the Hatchet worker by running the following command:

poetry run hatchet

Triggering a workflow

Follow the instructions in the root project setup to launch the playground and start workflow runs.

Project Overview

Example Workflows

The project contains example workflows in the ./src directory. These workflows are registered with hatchet in ./src/main.py which is started when running poetry run hatchet.

Super Simple Workflows

The project includes a variety of basic workflows to demonstrate Hatchet's core capabilities, each showcasing different features:

1. Simple Workflow: Demonstrates a straightforward process flow, showcasing the basics of setting up a workflow in Hatchet.
2. Async Workflow: An example of using async def together with asyncio.sleep.
3. Concurrency Limit Workflow: Shows how to manage concurrency limits within workflows to ensure that only a certain number of instances run simultaneously.
4. Directed Acyclic Graph (DAG) Workflow: Illustrates setting up workflows with dependencies that form a Directed Acyclic Graph, demonstrating the advanced orchestration capabilities of Hatchet.
5. Manual Trigger Workflow: Explains how to initiate workflows manually, offering control over workflow execution triggers.
6. Timeout Workflow: Demonstrates handling timeout scenarios within workflows, ensuring that long-running or stalled processes are appropriately managed.

Generative AI Workflows

For more complex use cases, the project includes examples that integrate with OpenAI's API for generative tasks:

1. Simple Response Generation: A single-step workflow that makes a request to OpenAI, showcasing how to incorporate AI services into Hatchet workflows.
2. Basic Retrieval Augmented Generation (BasicRag): A multi-step workflow that involves loading website content with Beautiful Soup, reasoning about the information, and generating a response with OpenAI, demonstrating the potential for complex, AI-driven processes.

Exposing the workflows via FastAPI

For a more complete example of how you might use Hatchet as part of a deployed production service, check out the FastAPI Example