Senior Data Scientist, Deep Recommender Systems

Senior Data Scientist, Deep Recommender Systems

Summary:

We are seeking a seasoned and hands-on Senior Data Scientist to lead the design, development, and deployment of our next-generation deep learning recommender systems. The ideal candidate will be a technical expert with a proven track record of building and optimizing large-scale recommender systems in a production environment. You will play a pivotal role in driving our business metrics by delivering highly personalized and relevant recommendations to our users.

Responsibilities:

• Lead the end-to-end development of deep learning-based recommender systems, from ideation and research to production deployment and monitoring.
• Design and implement multi-stage recommender system architectures, including efficient retrieval (candidate generation) and sophisticated ranking models.
• Develop and optimize deep learning models for recommendations using architectures such as Two-Tower, Wide & Deep, and Transformer-based models.
• Manage and create large-scale embedding layers for categorical features related to both users and products to effectively represent high-dimensional data.

• Utilize and evaluate various specialized libraries and frameworks likeTorchRec,TensorFlow Recommenders (TFRS),Microsoft Rec,Alibaba RecSysandNVIDIA Merlinto accelerate development and deployment.

• Work with large, sparse datasets and tackle unique challenges in recommender systems, such as the cold-start problem and the long-tail problem.
• Collaborate with data engineers, machine learning engineers, and product managers to ensure the entire recommendation pipeline is scalable, robust, and aligned with business goals.
• Design and execute A/B tests to rigorously measure the business impact of new models and features.
• Implement MLOps best practices for continuous integration, deployment, and monitoring of recommender systems in a production setting.
• Mentor junior data scientists and contribute to a culture of technical excellence and innovation.

Required Qualifications:

• Hands-On Experience is a Must:

  Proven experience building, deploying, and maintaining deep learning recommender systems at scale. ( 5M users and 3M products - india wide streaming platforms / india wide apps )

• Deep Learning Expertise:

  Extensive experience with deep learning frameworks such as

  PyTorch or TensorFlow

  .

• Recommender System Architectures:

  Deep understanding and practical experience with architectures like:

• Two-Tower Models:

  For efficient retrieval and candidate generation by learning separate embeddings for users and items.

• Wide & Deep Models:

  For balancing the memorization of specific feature interactions with the

  generalization

  of a deep neural network.

• Session-based and Sequential Models:

  For capturing the temporal dependencies and order of user behavior within a session or sequence of interactions.

• Collaborative Filtering Architectures:

  For leveraging deep learning to model user-item interactions, such as in

  Neural Collaborative Filtering (NCF)

  .

• Graph Neural Networks (GNNs):

  For learning powerful item and user embeddings by representing interactions as a graph and propagating information.
• Embedding Layers

  :

  For handling sparse, high-dimensional categorical features and creating dense vector representations at scale.

• Specialized Libraries:

  Hands-on experience with at least two of the following libraries or frameworks:

• TorchRec

• TensorFlow Recommenders (TFRS)

• NVIDIA Merlin

• Microsoft Recommenders

• NVIDIA Recommenders

• Alibaba Recommenders

• Scalability & MLOps:

  Demonstrated ability to build and deploy systems that handle terabytes of data with low latency. Experience with distributed training and real-time serving is highly valued.

• Problem-Solving:

  A strong track record of solving common recommender system challenges such as the cold-start and long-tail problems.

• Analytical Skills:

  Strong knowledge of experimental design and A/B testing methodologies.

• Programming Proficiency:

  Expert-level proficiency in Python and its data science ecosystem (Pandas, NumPy).

Preferred Qualifications:

• Advanced degree (M.S. or Ph.D.) in a quantitative field such as Computer Science, Statistics, or a related discipline.
• 3 year experience building recommendations systems at scale
• Experience with big data technologies (e.g., Spark, HDFS, Databricks) and vector databases.
• Familiarity with other deep learning architectures for sequence modeling (e.g., Transformers).