Note: TaskNode is currently in alpha. While fully functional, you may encounter occasional issues. Please report any bugs on our GitHub issues page.
Do you ever:
- Keep your computer on all night just to run a script?
- Run scripts that would benefit from a much faster internet connection?
- Have too many scripts running at a time?
TaskNode is a powerful command-line tool that lets you run Python scripts asynchronously in the cloud with zero infrastructure setup. Submit a task, and we'll handle the rest.
- Zero Configuration: Just install and run - we handle all the cloud setup
- Dependency Management: Automatic detection and packaging of project dependencies
- Cross-Platform: Works on Linux, macOS, and Windows
- Notifications: Get an email when your task is complete
First, install TaskNode:
pip install tasknode
Optionally, generate a sample Jupyter notebook to run:
tasknode generate-sample # copies sample.ipynb to current directory
Then, submit a Jupyter notebook to run in the cloud:
tasknode submit sample.ipynb
# or
tasknode submit your_script.py
TaskNode also supports Jupyter notebooks. Just submit your notebook (e.g., tasknode submit notebook.ipynb
), and we'll convert it to a Python script and run it for you.
tasknode help
When you submit a script, TaskNode automatically:
- 📁 Packages your project directory
- 🔍 Excludes development folders (.git, venv, pycache, etc.)
- 📝 Captures dependencies in requirements-tasknode.txt
- ℹ️ Records Python version and system information
- 🔒 Securely uploads everything to our cloud infrastructure
TaskNode securely uploads your files to our cloud infrastructure and deletes them after your task is complete. If the task fails, the files are retained for 24 hours in case you need to re-run the task.
Any outputs generated are retained for up to 72 hours, during which time you can download them.
TaskNode automatically detects your dependencies and packages them in a separate file called requirements-tasknode.txt
. This file is uploaded to our cloud infrastructure and installed when your task is run.
Right now, TaskNode runs on Docker containers with Python 3.12. Most Python 3 code will work without any changes, but we're working on adding support for other Python versions.
All data is processed in the AWS US East (N. Virginia) region.