Ayushi Singh Provides a Glimpse of Netflix's Strategic Use of Real-Time Data

Ayushi Singh Provides a Glimpse of Netflix's Strategic Use of Real-Time Data

Ayushi Singh, Senior Software Engineer at Netflix

Video preview
Ayushi Singh
Ayushi Singh
Senior Software Engineer at Netflix

Ayushi Singh is a Senior Software Engineer in the Online Datastores team at Netflix. With her expertise in open-source projects like Apache Cassandra, Ayushi is helping drive Netflix's vision to entertain the world through scalable data solutions. Prior to her role at Netflix, Ayushi was a member of the Core Platform and Infrastructure team at PayPal implementing cloud platforms in a hybrid cloud setting. Her contributions empowered PayPal's business teams to scale their applications, serving millions of customers globally and processing billions of dollars in payments.

In this Digital Champions interview, Ayushi emphasizes the importance of creating an inclusive space for learning and collaboration in the technology industry, especially for women. Her insights are invaluable to fostering diversity in the field. Ayushi also discusses the power of Apache Cassandra and provides a glimpse into Netflix's strategic use of real-time data for content creation, infrastructure monitoring, and marketing strategies.


I’m Ayushi Singh. I work as a senior software engineer at Netflix. I'm a part of the data platform organization. Being a Digital Champion for me means sharing my experiences and learning with a broader audience. I really want to encourage others, especially women, to make use of technology and want to enhance on the fact that everyone has the skills to learn it and once they learn it, they can enhance their life in so many different ways. It helps in creating that inclusive space where we can learn, collaborate and grow together.

Netflix streaming is built on Cassandra and we are moving to 4.1 version of Cassandra which will help us in eliminating multiple challenges which we face with Cassandra 3.0. It will help us scale faster. Because of the zero copy streaming, nodes will be able to join the cluster faster and we can scale the cluster horizontally in a few hours.

Real-time data can help Netflix in so many ways, but I want to highlight four points. First is a better recommendation system. The second is basically better content creation. The third is we can have better infrastructure monitoring and relocating the resources at the right places and the fourth is we can have better marketing strategies for it.

The next major step for Netflix in our real time data journey is gaming. With gaming, it brings a  lot of different challenges than streaming. Because the way we used to handle data for streaming, it will be different the way we will be handling it for gaming. We have to process data differently. We have to model it differently. We have to we have to find a correlation between the data and how can we make it more efficient for our end user. I feel that with Cassandra 5.0 transactions, we will be really able to leverage it, and we'll be able to enhance our gaming for the end users.