Software Engineer 4/5 – Data and Feature Infrastructure, AI Platform

Netflix•,

76d•$466,000 - $750,000

About The Position

Machine Learning/Artificial Intelligence powers innovation in all areas of the business, from helping members choose the right title for them through personalization, to better understanding our audience and our content slate, to optimizing our payment processing and other revenue-focused initiatives. Building highly scalable and differentiated ML infrastructure is key to accelerating this innovation. ML models can only be as good as the data we provide them. That's why we continue to innovate on making data and feature engineering as simple, scalable, and efficient as possible. In this role, you will have the opportunity to build a next-generation ML data and feature platform to significantly improve the productivity of ML practitioners. Our goal is to enable our ML practitioners to easily define and test ML features and labels, while our platform takes care of the computation, storage, and serving of feature values for both high-throughput training and low-latency member-scale inference use cases. You will also have the opportunity to build a centralized feature and embedding store to enable sharing across various ML domains. Unlocking access to these shared datasets will foster innovation through ML in new business areas that otherwise wouldn’t have been feasible. You will collaborate closely with ML practitioners and domain experts to ensure that our models are built with high-quality features and labels. You will also get to work with the broader AI Platform organization to deliver a cohesive end-user experience that significantly improves the productivity of ML practitioners.

Requirements

Experience in building ML or data infrastructure
Strong empathy and passion for providing a fantastic user experience to ML practitioners
Experience in building and operating 24/7 high-traffic and low-latency online applications
Experience with large-scale data processing frameworks such as Spark, Flink, and Kafka
Experience with data storage and serving technologies such as Iceberg and Cassandra
Experience in working with and optimizing Java and/or Python codebases
Experience with public clouds, especially AWS
Self-driven and highly motivated team player

Nice To Haves

Experience in building and operating ML feature stores, such as Chronon
Experience in building embedding-based retrieval systems
Experience working with Notebooks such as Jupyter or Polynote

Responsibilities

Design and build a near-real-time feature computation engine to generate ML features for both high-throughput training and low-latency inference applications.
Operate and manage the feature computation pipelines and feature serving infrastructure for various ML models across multiple ML domains.
Build and scale systems that accelerate training through performant data loading, transformation, and writing.
Create frameworks to streamline and expedite the availability of new data for training and serving.
Develop feature stores that enable feature discovery and sharing.
Increase the productivity of ML practitioners by making it easy to define and access features and labels for experimentation and productization.