Skip to content

Getting started

Daisi ( is a scientific Cloud Computing platform running Python serverless functions, named Daisies.

With Daisi you can:

  • create web services, automate tasks, host apps… orders of magnitude easier and faster than traditional cloud platforms.
  • parallelize compute intensive workloads by scaling up instantly to thousands of processes on demand.
  • share your code and discover thousands of ready-to-call functions and apps !

Daisi brings the power of cloud computing into the hands of every developer, scientist, engineer by automatically deploying and creating endpoints for any function of a Python code, which can then be invoked seamlessly from any environment.

Any regular Python code can be turned into a Daisi by simply registering its Github repository in the Daisi platform, with no need to write any specific code.


This project is under active development

Quick start with the pydaisi Python package

Calling "Hello World !"

  1. Install pydaisi (requires Python 3.8+)

    pip install --upgrade pydaisi
  2. Call the "Print Hello" Daisi and invoke the hello() endpoint

    import pydaisi as pyd
    greeting_daisi = pyd.Daisi("exampledaisies/Print Hello")
    greeting_daisi.hello(name = "Daisi user").value

Check the code of this Daisi on "Print Hello".
And browse to more discover Daisies !

What just happened ? Cloud Computing made easy !

We have instantiated a Daisi object and passed the name of a serverless function from the Daisi platform in argument. This is how you can call any available Daisi. The pattern to call a Daisi is "username/daisiname", except if the Daisi is public and its name is unique, in which case you can omit "username".

A Daisi can have multiple endpoints. Endpoints are created automatically for every function in the main script of your code. They are simply invoked as a method of the Daisi object, with the method name being the name of the function. In this example, we have invoked the hello endpoint of the Print Hello Daisi.

With pydaisi you can send and receive arbitrary Python objects, as if you were passing arguments to a local function. No need to serialize / deserialize, pydaisi handles it automatically.

Invoking a Daisi endpoint returns a DaisiExecution object. The value returned can be simply retrieved by querying its .value attribute.


Public Daisies can be called without being authentified. Private Daisies shared within a team require authentication (see the Authentication section)

Daisies can also be apps !

You can give a nice front end to your Daisies, and the platform currently supports Streamlit (more frameworks to come). And convert your Daisi into a fully deployed app with an API for computer access....

Check this Daisi: "Print Hello App"