Staff Data Scientist

Walmart•Sunnyvale, CA

5d•$225,077 - $286,000•Onsite

About The Position

We are seeking seven (7) Staff Data Scientists to join our team. In this role, you will be responsible for designing, developing, and productionizing large-scale personalization and computer-vision systems. This includes areas such as candidate generation, ranking, visual similarity, and image-based recommendations. You will also design, implement, and support the deployment of large-scale data pipelines, collaborating with engineering teams to integrate models into serving systems. A key part of this role involves contributing to the ML/DL roadmap for personalization, leading selected phases of the product lifecycle, and collaborating on online-inference deployments. You will analyze pre/post-deployment metrics, recommend performance tuning changes, and adapt pretrained vision and multimodal models via transfer learning for retrieval, ranking, and relevance. Developing generative-modeling workflows for automated asset creation and contributing to metric definitions, experiment design, and A/B testing are also core responsibilities. You will present readouts with recommendations, contribute to architecture and road-mapping, and participate in invention disclosures and patents.

Requirements

Master’s degree or equivalent in Statistics, Economics, Analytics, Mathematics, Computer Science, Information Technology, Engineering (any), or related field and 2 years of experience in an analytics related field; OR Bachelor’s degree or equivalent in Statistics, Economics, Analytics, Mathematics, Computer Science, Information Technology, Engineering (any), or related field and 4 years of experience in an analytics related field.
Experience with coding in Python, SQL, Unix Shell, Git, NumPy, Pandas, and Matplotlib.
Experience with big-data tools Spark/PySpark and Scalding for large-scale processing and feature pipelines.
Experience with Computer Vision for working with existing images (segmentation/detection, visual similarity) and applying pretrained vision models/embeddings.
Experience implementing deep learning models in a modern framework (PyTorch, TensorFlow).
Experience with CNN-based image embeddings and multimodal (vision–language) models.
Experience applying pretrained models and transfer-learning techniques.
Experience with text embeddings and language models, including TF-IDF, Word2Vec, Doc2Vec, and RNN/LSTM.
Experience with machine-learning algorithms including classification and clustering (support vector machines) and dimensionality reduction (PCA).
Experience with recommender systems, including matrix factorization and collaborative filtering, and embedding-based retrieval and ranking.
Experience programming computer algorithms.
Experience contributing to statistical inference and hypothesis testing, including familiarity with offline relevance metrics (Recall@K, NDCG) and A/B methodology, and presenting experiment readouts.
Experience contributing to production Machine Learning (ML) systems, including batch/stream processing and integration with serving/online-inference systems in partnership with engineering.
Experience validating deployments and analyzing runtime performance.
Experience with generative modeling and foundation-model adaptation (GANs, VAEs, diffusion) and evaluating generated outputs.
Employer will accept any amount of graduate coursework, graduate research experience or professional experience with the required skills.

Responsibilities

Design, develop, and productionize large-scale personalization and computer-vision systems (candidate generation, ranking, visual similarity, image-based recommendations).
Design, implement, and support deployment of large-scale data pipelines; collaborate with engineering to integrate models into serving systems.
Contribute to discovery and formulation of the ML/DL roadmap for personalization and lead selected phases of the product lifecycle in partnership with Product and Engineering.
Collaborate on online-inference deployments, analyze pre/post-deployment metrics, and recommend performance/tuning changes.
Adapt pretrained vision and multimodal models via transfer learning for retrieval, ranking, and relevance; manage embeddings/feature stores.
Develop generative-modeling workflows for automated image/text asset creation (e.g., diffusion/GAN/VAE-style or equivalent) with quality/brand checks and measured impact.
Contribute input to metric definitions and experiment design; collaborate on A/B and offline relevance evaluation; present readouts with recommendations.
Contribute to architecture and road-mapping via design discussions and documentation; contribute to invention disclosures/patents.