Sr Applied Research Scientist - Recommendations, Apple Media Products
Santa Clara Valley (Cupertino), California, United States
Software and Services
- Experience with MapReduce strongly preferred.
- Familiarity with following programming tools preferred: Hadoop, Spark, Java, Scala, Python, R, Tensorflow, CUDA.
- Solid understanding of data mining, Big Data and statistical models.
- Large distributed systems and performance tuning experience.
- Experience with CPU architectures: Intel i386 and x86_64 family; ARM family.
- Knowledge of protocols and standards: Transport Stream file format; HTTP and HTTPS.
- Ability to handle high workload and multiple responsibilities.
- Strong written & oral communication skills.
Research, design and develop machine learning models for iTunes & App store recommendations. Propose, prototype and evaluate the algorithm improvements. Build personalized recommender systems for Apple Music, Apps & Games Recommendations, Video, Podcast and Books Recommendations using one or more of the following methods: Deep Learning, Matrix Factorization, Factorization Machines, Text Mining, NLP, Learn to Rank models etc. Build a pipeline for analyzing big data that consists of both content and user data on Hadoop using map/reduce techniques. Adapt machine learning algorithms to large scale data (big data). Develop and build cross validation for your models. Conduct human judgments and A/B experiments, improve ranking models based test data. Derive insights from the experimentation and convert them into feature improvements. Ship production quality code for the offline model building and work with engineering team to develop/deploy the run time system for the model. Analyze software performance problems and implement optimizations. Active contribution to identify areas of improvement in personalization and recommendation products. Ability to adapt latest in literature in the area to build efficient and scalable models.
Education & Experience
MS in Machine Learning or Statistics or related degree or equivalent experience. PhD preferred.