Staff, Data Scientist

Walmart · Retail · Sunnyvale, CA

The Cortex Team at Walmart is building the next generation of AI conversational platforms, focusing on personal assistants for customers accessible via voice, text, and multi-modal experiences. The role involves improving and evolving NLU services, considering conversational context and multi-modal interactions, and potentially building demos and proof-of-concepts. The position requires extensive experience with NLU/NLP, deep ML models (Transformers, LLMs), and deploying end-to-end Generative AI systems. Experience with classical ML, conversational data analysis, and taking projects from scoping to launch is also necessary. Preferred qualifications include experience with multimodal solutions, agentic systems, LLM serving optimizations, and multi-LoRa.

What you'd actually do

  1. building and designing the next generation of Natural Language Understanding (NLU) services that other teams can easily integrate and leverage, and build rich experiences: from pure voice and text shopping assistants (Siri, Sparky), to customer care channels, to mobile apps with rich, intertwined, multi-modal interaction modes (Me@Walmart).
  2. improve and evolve our capabilities, taking into account the full conversational context, multi-modal interactions, and an ever increasing list of use cases.
  3. building demos, proof of concepts, creating white papers, writing blogs, etc.
  4. designing, developing and deploying end-to-end Generative AI systems using transformer-based LLM architectures (both open source and close source)
  5. analyzing conversational data to identify patterns and conducting error/deviation analysis; passion for fixing issues in the data and finding the optimal representation.

Skills

Required

  • Master's degree or certification in Machine Learning, Computer Science, Engineering, Mathematics, Statistics or any other related field, with 3+ years work experience.
  • Extensive experience with NLU/NLP, more recent deep ML models (e.g., Transformers, BERT, Llama, GPTs, Gemini) and safe fine-tuning of these.
  • Extensive experience in designing, developing and deploying end-to-end Generative AI systems using transformer-based LLM architectures (both open source and close source)
  • Experience with Python; solid knowledge SQL.
  • Hands on experience with classical ML models, test/train/evaluation metrics, key parameters/techniques that affect model performance
  • Understanding of relevant statistical measures such as confidence intervals, significance of error measurements, development and evaluation data sets.
  • Ability to take a project from scoping requirements through actual launch.
  • A continuous drive to explore, improve, enhance, automate, and optimize models and products.
  • Experience analyzing conversational data to identify patterns and conducting error/deviation analysis; passion for fixing issues in the data and finding the optimal representation.
  • Excellent oral and written communication skills.

Nice to have

  • PhD in a relevant field (Machine Learning, Computer Science, Engineering, Mathematics, Physics, Statistics or a related field)
  • Experience developing multimodal solutions for Generative AI and related applications at scale
  • Exposure to real-world, production grade agentic systems.
  • Familiarity with LLMs serving optimizations and multi-LoRa
  • Ability to develop experimental and analytic plans for data modeling processes, use of strong baselines, ability to accurately determine cause and effect relations
  • Strong attention to detail and exceptional level of organization
  • Proven ability to achieve results in a fast paced, highly collaborative, dynamic work environment
  • Hands-on expertise in many disparate technologies and the full model lifecycle, typically ranging from data pipelines, data extraction, model training, model serving, labeling tools, ML-ops, ad-hoc tooling and all points in between.

What the JD emphasized

  • extensive experience with NLU/NLP, more recent deep ML models (e.g., Transformers, BERT, Llama, GPTs, Gemini) and safe fine-tuning of these.
  • extensive experience in designing, developing and deploying end-to-end Generative AI systems using transformer-based LLM architectures (both open source and close source)
  • Hands-on expertise in many disparate technologies and the full model lifecycle, typically ranging from data pipelines, data extraction, model training, model serving, labeling tools, ML-ops, ad-hoc tooling and all points in between.

Other signals

  • building and designing the next generation of NLU services
  • delivering the worlds best personal assistants to Walmart’s customers
  • powering the vision of delivering the worlds best personal assistants to Walmart’s customers, accessible via natural voice commands, text messages, rich UI interactions, and a mix of all of the above via multi-modal experiences