Senior Data Scientist — Alexa Shopping Analytics

Amazon.com Services LLC; Seattle (WA); On-site; USD 159,000 - 215,000; Full time

Benefits offered by this job

Health insurance

401(k) matching

Paid time off

RSUs

Parental leave

Sign-on payments

Job summary

Amazon.com Services LLC is seeking a Senior Data Scientist for Alexa For Shopping (Rufus) to lead multi-agent orchestration and a self-improving agent system. You will extract insights from unstructured data at scale and surface actionable metrics for product analytics.

The role requires deep experience with NLP pipelines, AI/ML tech, and end-to-end ownership of multi-agent system design, deployment, and continuous improvement in a production environment.

Qualifications

• 4+ years of data scientist experience.
• 5+ years with SQL, Python, or statistical software.
• Experience with multinomial logistic regression or similar models.
• 5+ years working with AI/ML, GenAI, analytics, databases or storage.
• Python proficiency for modeling and data manipulation (pandas, numpy, scipy).
• Experience extracting structured signals from unstructured text at scale (NLP).

Responsibilities

• Lead end-to-end multi-agent orchestration with significant product impact.
• Own multi-agent topology and inter-agent communication protocols.
• Design and manage context window strategy across agents.
• Own system prompts, routing prompts, and chain-of-thought scaffolding.
• Define metrics for factual grounding, hypothesis novelty, and reasoning quality.
• Oversee evaluation signal propagation into prompts and routing decisions.
• Oversee schema grounding, vector indexing, and domain-specific kNN queries.
• Own embedding strategy, intent classification accuracy, and entity extraction quality.

Skills

Data science experience

SQL

Python

NLP / unstructured data

Statistical modeling

AI/ML / GenAI

Pandas / NumPy / SciPy

Tools

AWS Bedrock

Strands SDK

LangGraph

CrewAI

AutoGen

SageMaker

Neptune

Neo4j

Job description

Amazon.com Services LLC is seeking a Senior Data Scientist for Alexa For Shopping (Rufus) to lead multi-agent orchestration and a self-improving agent system. You will extract insights from unstructured data at scale and surface actionable metrics for product analytics.

The role requires deep experience with NLP pipelines, AI/ML tech, and end-to-end ownership of multi-agent system design, deployment, and continuous improvement in a production environment.