(usa) Staff, Data Scientist

Walmart · Retail · ISD - DGTC AR BENTONVILLE

Staff Data Scientist role at Walmart focused on leveraging Computer Vision, Generative AI, and VLMs to build and deploy advanced retail solutions. Responsibilities include designing, training, and deploying computer vision models for object detection and personalization, building foundation model pipelines with multimodal signals, and mentoring team members on generative modeling techniques. The role also involves integrating models with MLOps for performance optimization and deployment in production environments.

What you'd actually do

  1. Design, train, and deploy computer vision models for object detection, image-based personalization, including product discovery, visual recommendations, and generative image content.
  2. Build and scale foundation model pipelines that leverage multimodal signals — including text, images, and customer embeddings — to power highly personalized visual experiences.
  3. Mentor team members on best practices, lead innovation efforts on image generative modeling, including diffusion models, CNNs, or text-to-image synthesis aligned with retail context.
  4. Design model workflow that integrates with MLOps to optimize model performance and latency, deploy and monitor models in production environments, ensuring performance and scalability.
  5. Collaborate with cross-functional teams to translate business requirements into data-driven solutions.

Skills

Required

  • MS or PhD in Computer Science, Machine Learning, Computer Vision, Applied Mathematics, or a related technical field
  • 5-8 years of industry or research experience
  • Deep learning for computer vision (CNNs, ViTs, diffusion models)
  • Building and deploying image generation, manipulation, or understanding systems in production
  • Python
  • PyTorch
  • TensorFlow
  • OpenCV
  • torchvision
  • Hugging Face
  • GPU deepstream
  • Multimodal models (CLIP, BLIP, Flamingo)

What the JD emphasized

  • expertise in data sourcing, model development, validation, and deployment within complex data ecosystems
  • Demonstrated expertise in deep learning for computer vision, including CNNs, vision transformers (ViTs), and diffusion-based generative models.
  • Experience building and deploying image generation, manipulation, or understanding systems in production — such as visual search, personalized image recommendations.

Other signals

  • computer vision
  • generative AI
  • deep learning
  • VLMs
  • multimodal signals
  • image generation