Netflix reveals the way it uses Python for operations management and analysis, in addition to cybersecurity and networks.
From their blog we can read: “We use Python throughout the content lifecycle, from deciding which content to fund to running the CDN that serves the final video to 148 million members. We use and contribute to many packages of Python open source.”
On the blog, the Netflix Pythonistas They also told us the way they used Python.
The “Demand Engineering” department is responsible for allocating and distributing resources depending on traffic and failover. “We are proud to say that our team's tools are primarily built in Python,” they tell us on their blog. You will be able to know all these tools used and apply them in our master in Python programming.
What tools do you use?
Uses Numpy and Scipy to do numerical analysis, Boto3 to make changes to the AWS infrastructure, rq to run asynchronous workloads and all under one Flask API.
The same mathematical libraries, together with Ruptures either pandas They are used by the Netflix CORE team for analytical work on alerts and statistics.
Python is also used for tasks data automation, exploration and cleansing, and as a source for visualization work. Additionally, it is essential in the area of personalization machine learning, recommendation algorithms, illustration personalization, and marketing algorithms.
Due to the evolution of the platform and they constantly test new approaches, they use Jupyter Notebooks to drive many of their tests.
Netflix usa Python practically in all areas of the company, applies machine learning to hundreds of use cases across all divisions of the company, most of those applications powered by Metaflow, a Python framework that makes it easy to run ML projects from the prototype stage to production.
Metaflow As indicated in the blog, it takes Python to its limits, making the most of it.