Principal Data Scientist at Comcast (Philadelphia, PA)

Comcast's Technology & Product organization works at the intersection of media and technology. Our innovative teams are continually developing and delivering products that transform the customer experience. From creating apps like TVGo to new features such as the Talking Guide on the X1 platform, we work every day to make a positive impact through innovation in the pursuit of building amazing products that are enjoyable, easy to use and accessible across all platforms. The team also develops and supports our evolving network architecture, including next-generation consumer systems and technologies, infrastructure and engineering, network integration and management tools, and technical standards.

Job Description

Bring a combination of mathematical rigor and innovative algorithm design to create recipes that extract relevant insights from billions of rows of data to meaningfully improve Comcast user experience.

Are you passionate about network, digital media, entertainment, and software services? Do you likebigchallenges and working within ahighly-motivatedteam environment?As a Data Scientist, you will research, develop,support and apply new techniques using real-time distributed computing architectures.You willemploy your skills to deliver insights into customer and network behavior to drive business decisions that will shape the future of Comcast and be part of a team that thrives onbigchallenges, results, quality, and agility.

Who does the Data Scientist work with?

The Data Engineering and Science team as part of our Next Generation Access Network is a diverse collection of professionals who work with a variety of organizations ranging from software engineering teams whose software integrates with analytics services, network architect and engineers, to service delivery engineers who provide support for our product, testers, operational stakeholders with all manner of information needs,and executives who rely ondatafor data based decision making.

What are some interesting problems you'll be working on?

Develop models capable of processingmillions of events per second and multi-billions of events per day, providing both a real time and historical view into the operation of our products and services.Work on high performance real time data stores and a massive historical data sets usingbest-of-breed and industry leading technology.Work closely with various engineering teams to solve key optimization, insight and access network data challenges.

Where can you make an impact?

The Comcast Next Generation Access Network Data Engineering and Science teamis acquiring, studying, simulating, and modeling to enable data as a key driver and core functional component toward better understanding, predicting, and dynamically optimizing the core network to improve overall user experience. Success in this role is best enabled by a broad mix of skills and interests ranging from traditional distributed systems software engineering prowess to the multidisciplinary field of data science.

Responsibilities:

• Building a strong intuitive understanding of the problem domain (Next Generation Access Networks). Identify testable hypotheses to explain interesting phenomena in the domain.
• Selectingand transformingfeatures, building and optimizing classifiers using machine learning techniques

• Integrating data from multiple sources including third party sources.

• Data mining using state-of-the-art methods

• Enhancing data collection procedures to include information that is relevant for building analytic systems
• Frequent meeting/communication with clients to interpret their needs, plan/organize, and discuss progress and results
• Developing actionable quantitative models in the areas of effectiveness, ROI, pricing and optimization.

• Doing ad-hoc analysis and presenting results in a clear manner

• Creating automated anomaly detection systems and constant tracking of its performance

• Creating automated evaluation environment of complex models and constant tracking of relevant performance
• Develop and communicate goals, strategies, tactics, project plans, timelines, and key performance metrics to reach goals
• Review, direct, guide, inspire the analytical work of more junior staff

Here are some of the specific technologies we use:

• Spark(AWS EMR), AWS Lambda
• SparkStreaming and Batch
• Avro, Parquet
• Kafka
• MemSQL, Cassandra,HBase,MongoDB, RDBMS
• Caching Frameworks(ElasticCache)
• Elasticsearch, Beats, Logstash, Kibana
• Java, Scala, Go, Python, R
• Git,Maven, Gradle, Jenkins
• Rancher, Puppet, Docker, Ansible, Kubernetes
• Linux
• Hadoop (HDFS, YARN, ZooKeeper, Hive), Presto
• Keras, TensorFlow

Skills & Requirements:

• Graduate degree or PHD in the following areas: Statistics, Data Science, Computer Science or Operations Research.
• 6+ years working within an enterprise data lake/warehouse environment or big data architecture
• Excellent understanding of machine learningtechniques andalgorithms, especially in the deep learning area-- both theoretical underpinnings and craft (Systems such as Tensorflow, Theano, Caffe, scikit.learn and their APIs).

• Excellentapplied statistics skillsand understanding of probabilitydistributions, statistical testing, regression, etc.

• Experience with common data science toolkits, such asscikit-learn, R, etc. Excellence in at least one of these is highly desirable.

• Great communication skills.

• Experience with data visualization tools, such as D3.js, GGplot,Matplotlib,etc.

• Proficiency in using query languages such as SQL and Hive.

• Experience with NoSQL databases, such asRedis/ElasticCache,Cassandra, HBase

• Good scripting and programming skills, such as Java, Scala, R, Python, or Spark
• Data-oriented personality

Comcast is an EOE/Veterans/Disabled/LGBT employer

by via developer jobs - Stack Overflow