Netflix is the world's leading internet streaming service with over 209 million members in 190 countries. Our members enjoy a wide variety of streamed video per day, using over 1000 types of devices, generating more than a third of downstream Internet traffic in North America. The Content Knowledge Graph team within Data Science & Engineering is based in the entertainment capital of the world and is passionate about providing our colleagues with rich data and insights. Our charter is to map the world's content. Data and insights from the CKG enable teams to rapidly experiment and drive business decision-making such as predicting the value of content, forecasting global demand, and analytical insights into key drivers of a movie's success. We are looking for candidates looking to drive the next wave of innovation and business impact which requires a blend of creative thinking and applied AI/ML research within Machine Learning, Deep Learning, Graph Learning & Analytics, and Information Retrieval using Knowledge Graphs and web-source data.
As a Senior Applied ML Scientist in the CKG team, you will use world-class engineering and AI/ML techniques on real-world, big data to directly impact the evolution of our content catalog. You will have the opportunity and unlimited scope to prototype, test, and execute cutting-edge research in a variety of topics (examples include but are not limited to graph learning and analytics, information retrieval, NLP, and personalization) while working cross-functionally with other scientists and engineering across a wide range of stakeholders: content analytics, valuation, demand modeling, and product personalization teams leveraging our Knowledge Graph. You will be actively engaged with the Netflix and external ML community through publications, organizational activities, and academic collaborations.
We are looking for an experienced researcher that can strategize, roadmap, and execute a solid research agenda, get others behind their ideas and foster an open environment of collaboration, innovation and intellectual excitement.
Challenges you'll tackle:
• You will research and build robust ML/DL models that improve entity extraction, classification, resolution, and disambiguation within the CKG across multiple languages (e.g. English, Korean, etc.).
• You will lead efforts to quantify the performance of our entity extraction/matching algorithms by measuring the quality and health of entities/relations within CKG; and partner with data specialists, engineers, and our labeling workforce to build a continual learning ML and quality assurance framework.
• You will partner with internal scientists to develop graph analytics and ML models capable of answering questions like ''how does talent and IP to determine a movie's likelihood of success and creative excellence and what can we learn from various data modalities (graph, text, images/video)?''.
• You will work closely with engineers on detailed requirements, technical designs, and implementation of end-to-end inference solutions at ''Netflix scale'' and help drive tooling for graph analysis, model development, testing and serving.
• Ample ML research experience, in academia and industry, in fields such as graph learning, content understanding (NLP, NLU, CV), personalization/recommendation systems
• A track record of ideating and leading ambitious research projects inspired by real-world challenges resulting in tangible business impact; we also love peer-reviewed publications, conferences, blogs.
• Passion for building strong relationships with stakeholders and colleagues to tackle big, cross-functional problems
• Exceptional communication with technical and non-technical audiences
• Comfort with ambiguity; ability to thrive with minimal oversight and process
• You excel in at least one major language (e.g., Python) and ML/DL framework (e.g. PyTorch, TensorFlow, Keras).
• Bonus: experience with SQL (any variant), big-data tech (e.g.,Spark, Hadoop, Hive) for ML pipelines, familiarity with graph databases (Neptune, Neo4J, TigerGraph), graph query languages (Gremlin, SPARQL, Cypher), or graph data models (RDF, Property)
Jobcode: Reference SBJ-gxomvq-52-205-167-104-42 in your application.