Skip to main content

Senior Data Scientist, Prime Video Playback Analytics

Job ID: 1881201 | Amazon Dev Centre Canada ULC

DESCRIPTION

Job summary
Shape the future of Prime Video Playback Analytics by applying your data science skills to our ocean of playback telemetry to deliver new insights into data quality and performance.

Playback Analytics manages the telemetry from hundreds of device types and millions of Prime Video customers worldwide generating trillions of events. Our data pipelines are amongst the largest in the world of video and our team delivers real-time data and tooling for a diverse group of internal customers.

Our data is used to monitor and improve the quality of playback in real-time, feed anomaly detection systems, support customer troubleshooting and provide a range of automated reports supporting dozens of business units within Prime.

This role will own the creation of new data quality tooling to measure and report on the integrity of our data across hundreds of dimensions and lead the design of solutions that surface automated insights into the data for our customers.

Key job responsibilities

  • You will design and implement code in Python (or R, Scala etc) to solve ambiguous and complex business problems where you will shape the direction of our data quality and insights tool chains.
  • You will work with software development engineers to turn your concepts into fully production ready solutions operating at scale.
  • You will analyze our data using a range of statistical or ML based approaches and explore the problem domain to evaluate possible approaches that yield the most accurate and actionable outputs.
  • You will lead our data science projects from concept to completion, including data gathering, experimentation, stakeholder management, technical documentation writing, testing and implementation.
  • You obsess over data quality, understand the impact of flawed data to our business and are able to combine mathematical, statistical, probabilistic and where appropriate, ML approaches to our problem domain.

A day in the life
Gaining a deep understanding of our data you will initially explore, experiment and deliver data science solutions that help quantify and surface actionable data quality issues. Solving our problems requires analysis of the syntax, semantics and sequences of telemetry and experimentation is core to success. For problems where you have identified the path forwards, you will be writing the critical path code that powers the end to end solution in collaboration with software development engineers as needed. Your contributions will shape the next generation of our data insights solutions which you will lead in collaboration with our senior engineers.

About the team
Playback Analytics comprises multiple teams that own the data platform, the streaming metrics computation engine, a range of published datasets consumed by diverse business units and the customer facing tooling for accessing petabytes of data in a readily consumable format. We support a range of customers such as playback engineers looking at diagnostic telemetry, partner teams working on algorithms for improving playback performance and operators monitoring live international sports events. We have recently completed a ground up re-architecture of our telemetry systems which sets the foundation for the new insights we can deliver as we move up the information pyramid.

BASIC QUALIFICATIONS

  • Bachelor's Degree
  • 5+ years of experience with data scripting languages (e.g SQL, Python, R etc.) or statistical/mathematical software (e.g. R, SAS, or Matlab)
  • 4+ years working as a Data Scientist

  • Located in a timezone no more than 3 hours outside of US Pacific time
  • Occasional travel may be required to Seattle for project kick-off or other key collaborations - subject to prevailing travel guidance.

PREFERRED QUALIFICATIONS

  • Familiarity with video telemetry is helpful but not required
  • Ability to independently solve problems in ambiguous domains
  • Experience with comparative statistical methodology
  • Machine Learning experience may be useful but is not a requirement




Amazon is committed to a diverse and inclusive workplace. Amazon is an equal opportunity employer and does not discriminate on the basis of race, national origin, gender, gender identity, sexual orientation, disability, age, or other legally protected status. If you would like to request an accommodation, please notify your Recruiter.