(usa) Senior, Data Scientist

Walmart Walmart · Retail · Sunnyvale, CA

Senior Data Scientist at Walmart focused on scaling and optimizing GPU infrastructure for model training, deploying generative LLM models for online inference, and developing foundation models for the e-commerce platform. The role involves benchmarking, distributed training, and MLOps best practices.

What you'd actually do

  1. Identify and source relevant data from multiple business domains, ensuring data quality and suitability for analysis.
  2. Develop, test, and deploy advanced analytical models using programming languages such as Python and SQL.
  3. Benchmark profiling tools to analyze and optimize GPU utilization and training throughput for machine learning workloads
  4. Implement distributed inference techniques for large-scale LLM deployments using VLLM
  5. Perform exploratory data analysis and feature engineering to enhance model accuracy and business insights.

Skills

Required

  • Python
  • SQL
  • distributed training techniques
  • data parallelism
  • model parallelism
  • APIs for model inference
  • scalable microservices design
  • infrastructure tooling
  • data visualization tools
  • data source identification
  • data quality assessment
  • data strategy

Nice to have

  • Horovod
  • DeepSpeed
  • PyTorch Distributed
  • mentoring associates
  • collaborating cross-functionally

What the JD emphasized

  • GPU utilization
  • training throughput
  • distributed inference
  • large-scale LLM deployments
  • VLLM
  • distributed training techniques
  • data parallelism
  • model parallelism

Other signals

  • scaling GPU infrastructure for model training
  • deploying generative LLM models for online inference
  • foundation model tailored for Walmart’s e-commerce platform
  • distributed inference techniques for large-scale LLM deployments
  • MLOps best practices