Applied Scientist , Amazon

Amazon Amazon · Big Tech · Palo Alto, CA · Applied Science

Applied Scientist role at Amazon focusing on improving shopping experiences using LLMs. The role involves post-training of LLMs, including instruction tuning, reward modeling, and reinforcement learning. Responsibilities include designing and running large-scale experiments, analyzing model behavior, and developing new training recipes to enhance capabilities like reasoning and user experience. Requires a PhD or Master's with significant experience, practical LLM experience, and a strong publication record.

What you'd actually do

  1. design and run large-scale experiments
  2. analyze model behavior
  3. develop new training recipes that directly improve core capabilities like reasoning, user experience, and other frontier paradigms that redefine what LLMs can do
  4. build the next generation of large language models
  5. work with talented scientists, engineers, and technical program managers (TPM) to innovate on behalf of our customers

Skills

Required

  • building models for business application
  • PhD, or Master's degree and 4+ years of CS, CE, ML or related field experience
  • Java, C++, Python or related language
  • LLM pre-training or post-training
  • working with organic, synthetic, agentic, or reasoning data for LLMs
  • Published research in top conferences (e.g., NeurIPS, ICML, ICLR, ACL, EMNLP, NAACL) and/or demonstrated significant industry influence in the field of AI

Nice to have

  • Unix/Linux
  • professional software development
  • SOTA LLMs
  • Multiple first-author LLM-related publications in top-tier conferences (e.g., NeurIPS, ICLR, ACL, EMNLP, NAACL)

What the JD emphasized

  • PhD, or Master's degree and 4+ years of CS, CE, ML or related field experience
  • Practical experience with LLM pre-training or post-training, and experience working with organic, synthetic, agentic, or reasoning data for LLMs
  • Published research in top conferences (e.g., NeurIPS, ICML, ICLR, ACL, EMNLP, NAACL) and/or demonstrated significant industry influence in the field of AI

Other signals

  • LLM post-training
  • instruction tuning
  • reward modeling
  • reinforcement learning
  • large-scale experiments
  • model behavior analysis
  • new training recipes
  • reasoning
  • user experience
  • frontier paradigms