Develop. Deploy. Monitor.

Curiosity is Sterblue's AI software system that uses the state of the art frameworks and libraries to get best results.

Completely open source

Why Sterblue Curiosity?

The reasons why we built and share this tool.

Cloud Native

Run multiple Experiments at the same time completly online. Logs and results are stored in the cloud with the proper versionning for easy reproduction.

Simple

Building on best practices with standard tools, you will not have trouble finding engineers that master the technologies of this stack, and you'll always find help thanks to large communities.

End-to-End

Automatically takes care of all the boilerplate around training a neural network, you only need to set your experiment parameters and it will handle the whole pipeline from data generation to deploying in production.

-50%

Infrastructure Costs

100%

Extensibility

5 Minutes

Experiment Time

How Sterblue Curiosity works

A tool built to allow you to jump from experimentation to deployment without obstacle.

1
.
Data

Define your dataset

Using a simple config file, Define the dataset used for training.
|
2
.
Train

Train your model

Once your dataset is defined, you can start a training online.
3
.
Deploy

Deploy your model

Deploy the model.

Open source state of the art Boilerplate

Sterblue Curiosity is based only on fully open source tools that represent the state of the art for Cloud Native Infrastructure.