Amazon Music is an immersive audio entertainment service that deepens connections between fans, artists, and creators. From personalized music playlists to exclusive podcasts, concert livestreams to artist merch, Amazon Music is innovating at some of the most exciting intersections of music and culture. We offer experiences that serve all listeners with our different tiers of service: Prime members get access to all the music in shuffle mode, and top ad-free podcasts, included with their membership; customers can upgrade to Amazon Music Unlimited for unlimited, on-demand access to 100 million songs, including millions in HD, Ultra HD, and spatial audio; and anyone can listen for free by downloading the Amazon Music app or via Alexa-enabled devices. Join us for the opportunity to influence how Amazon Music engages fans, artists, and creators on a global scale. Learn more at https://www.amazon.com/music.
The Data, Insights, Science and Optimization, Consumer Product and Tech (DISCO CPT) team is looking for a Data Scientist to join a team of Data Scientists, Business Intelligence Engineers and Data Engineers who analyze big data, provide analytics and insights as well as build models and algorithms that power the Music product experiences. DISCO CPT team focuses on accelerating Amazon Music customer growth by empowering Product teams to make sound, customer-centric decisions through data and insights. We build data pipelines, self-service analytics, insights and predictive models enabling acquisition, engagement and retention at scale with personalized customer touchpoints. In this role, you will set the science vision and direction for the team and collaborate with internal stakeholders across marketing, growth, product, science and finance teams to scale and advance our science offerings. The ideal candidate must be willing to effectively lead large scale science solutions, prioritize across multiple stakeholders and projects and be ready to jump into a fast-paced, dynamic and fun environment.
Key job responsibilities
**Predictive Modeling:**
* Develop predictive models to identify potentially fraudulent streaming activities, or content uploads. Utilize anomaly detection techniques to surface suspicious patterns in user behavior, and content metadata
* Collaborate with the FCRM team to refine fraud detection models and improve the accuracy of fraud identification in addition to improving streaming fraud entitlement estimation by using ML output to identify outliers
* Develop models to automate metadata enrichment and quality assurance processes
* Develop models to determine drivers of key performance metrics like MAU and HPC and automate the process of deep diving into variances in these metrics
**Experimentation and A/B Testing:**
* Collaborate with product and engineering teams to design rigorous experiments to evaluate the impact of new features or algorithms (e.g., the Playlist Song Recommendation experiments).
* Analyze the results of these experiments, identify learning, and provide recommendations to optimize the solutions
* Partner with Sr. Data Scientists to analyze and propose key experimentation success and guardrail metrics for business like AMOR, Catalog quality etc.
**Deep dives and analysis:**
* Develop metrics and rules that help identify metadata tag defects and improve metadata quality and enable more precise Catalog tiering
* Analyze the current state of metadata, including genre coverage, artist disambiguation, and track-level information and propose strategies to improve the overall metadata quality and coverage across the catalog
* Determine relationship between metadata quality improvements and customer engagement to guide prioritization of ingestion and coverage
**Cross - Functional Collaboration:**
* Partner with product, catalog, and engineering teams to understand business challenges and translate them into analytical science initiatives
* Evangelize the use of experimentation and advanced analytics science across the organization
* Mentor and train other analysts and data scientists to build their skills
