Sr. Applied Scientist, C360

Amazon Amazon · Big Tech · Seattle, WA · Applied Science

Senior Applied Scientist role focused on advancing Information Retrieval, NLP, and Large Language Models for e-commerce personalization. The role involves post-training LLMs (instruction tuning, reward modeling, RL, multi-modal alignment), designing large-scale experiments, analyzing model behavior, and developing training recipes to improve capabilities like reasoning and personalization. It also includes owning the scientific roadmap, leading end-to-end systems, driving technical decisions, mentoring, and publishing research.

What you'd actually do

  1. Own the scientific roadmap for personalization initiatives, identifying high-impact research directions and translating ambiguous business problems into well-defined ML formulations
  2. Design and lead end-to-end systems spanning recommendations, information retrieval, and LLM fine-tuning, from problem framing through offline experimentation to production A/B testing and launch
  3. Drive technical decisions on model architecture, training methodology, and evaluation frameworks, balancing scientific rigor with business impact and operational constraints
  4. Mentor and raise the bar for the science team through design reviews, paper discussions, and establishing best practices for experimentation and reproducibility
  5. Influence cross-functional strategy by partnering with engineering, product, and leadership to define the product vision informed by what's technically feasible and scientifically novel

Skills

Required

  • building machine learning models for business application experience
  • PhD, or Master's degree and 6+ years of applied research experience
  • Experience programming in Java, C++, Python or related language
  • Experience with neural deep learning methods and machine learning

Nice to have

  • Experience with modeling tools such as R, scikit-learn, Spark MLLib, MxNet, Tensorflow, numpy, scipy etc.
  • Experience with large scale distributed systems such as Hadoop, Spark etc.

What the JD emphasized

  • post-training
  • instruction tuning
  • reward modeling
  • reinforcement learning
  • multi-modal alignment
  • large-scale experiments
  • model behavior
  • training recipes
  • reasoning
  • personalization
  • frontier paradigms
  • scientific roadmap
  • recommendations
  • information retrieval
  • LLM fine-tuning
  • offline experimentation
  • production A/B testing
  • model architecture
  • training methodology
  • evaluation frameworks
  • scientific rigor
  • business impact
  • operational constraints
  • design reviews
  • paper discussions
  • experimentation
  • reproducibility
  • cross-functional strategy
  • engineering
  • product
  • leadership
  • product vision
  • technically feasible
  • scientifically novel
  • Publish
  • advance the state of the art
  • patents
  • publications
  • external engagement
  • conferences
  • large amounts of data
  • generate insights
  • execute experiments
  • statistical and ML models
  • business needs
  • collaboration
  • Scientists
  • Engineers
  • Product Managers
  • clear communication skills
  • innovate on behalf of the customer
  • strategically build features
  • mentor junior members
  • grow
  • innovations
  • safe place to try, fail and learn
  • culture of continuous improvement
  • leader and owner
  • creative space
  • entrepreneurial work environment
  • customer obsession
  • building machine learning models for business application experience
  • applied research experience
  • neural deep learning methods
  • machine learning
  • modeling tools
  • large scale distributed systems

Other signals

  • LLM fine-tuning
  • reward modeling
  • reinforcement learning
  • multi-modal alignment
  • large-scale experiments
  • personalization