Staff Data Scientist - Viewer experience

Job Summary: Lead the development of sophisticated recommendation algorithms that power personalized content discovery across JioStar's platform, driving measurable improvements in user engagement, satisfaction, and retention through data-driven innovations.
 
About the team: The Viewer experience team focuses on revolutionizing content discovery across JioStar's platform. Our aim is to provide users with an intuitive and personalized experience, enabling seamless exploration and enrichment of their interests. Our data science team develops advanced algorithms that understand user preferences and content relationships to deliver relevant recommendations at scale.
Working alongside ML engineers, developers, designers, and content curators, we translate business challenges into algorithmic solutions that analyze user preferences and behaviors to deliver immersive content discovery experiences. Our work directly influences how millions of users discover content daily, making our team a critical driver of user engagement and business success. We embrace a culture of experimentation, rigorous evaluation, and continuous improvement to enhance our recommendation capabilities.

Key responsibilities:

• Lead the vision and strategy for recommendation algorithms across the JioStar platform, identifying opportunities to enhance personalization and content discovery

• Design and develop sophisticated recommendation models leveraging collaborative filtering, content-based techniques, deep learning, and hybrid approaches

• Translate complex business requirements into data science solutions, driving alignment across product, engineering, and business stakeholders

• Build evaluation frameworks and metrics that measure recommendation quality across dimensions including relevance, diversity, freshness, and business impact

• Lead A/B testing and experimental design to validate algorithmic improvements and quantify business impact

• Develop novel approaches to recommendation challenges including cold-start problems, exploration-exploitation tradeoffs, and multi-objective optimization

• Collaborate closely with ML Engineering to ensure algorithms can be efficiently implemented at scale

• Analyze user behavior patterns to identify segments and personalization opportunities

• Provide mentorship to junior data scientists and establish best practices for the data science organization

• Stay current with research in recommendation systems and personalization, bringing innovative approaches to our platform

Skills and attributes for success:

• Deep expertise in recommendation system algorithms, including collaborative filtering, content-based, neural networks, and multi-stage approaches

• Experience with candidate generation, ranking, and slate optimization for personalized user experiences

• Strong background in reinforcement learning, bandits, and long-term reward modeling for recommendation systems

• Experience with transformer architectures, LLMs, and their application to personalization

• Knowledge of RLHF reward modeling/alignment techniques for improved recommendation systems

• Hands-on experience with Python, SQL, and TensorFlow/PyTorch for implementing and evaluating algorithms

• Knowledge of multi-task learning, transfer learning, and embedding techniques for users, items, and contexts

• Understanding of content life cycles, seasonality, and timing's impact on recommendation strategies

• Proven track record of developing recommendation systems that drive meaningful business outcomes

• Experience with experimental design and A/B testing methodologies for recommendation algorithms

• Ability to balance algorithmic exploration with user enjoyment in recommendation design

• Strong leadership capabilities with demonstrated experience mentoring junior data scientists

Preferred education and experience:

• Bachelor's in Computer Science, Statistics, Mathematics, or related quantitative field with 10+ years of experience in applied data science, including at least 5 years working specifically with recommendation systems.

• Experience in streaming media, entertainment, or similar content platforms strongly preferred.