LLM-powered Real-Time Context Scientist (Customer360)

Amazon; Seattle (WA); On-site; USD 143,000 - 193,000; Full time

Benefits offered by this job

Health insurance

RSUs

Job summary

Amazon is hiring an Applied Scientist to push the boundaries of LLMs, semantic retrieval, and customer understanding at scale. You will analyze large-scale behavioral data, design experiments, and ship production systems with millisecond latency, impacting customer experiences across Amazon’s surfaces.

You’ll work with cross-team partners (Alexa, Search, Recommendations) to define the technical roadmap, mentor others, and translate research into reliable, low-latency solutions that power real

Qualifications

• 3+ years building models for business applications.
• PhD or Master’s plus 4+ years in CS/CE/ML.
• Experience programming in Java, C++, Python or related language.
• Knowledge of algorithms, data structures, data mining and HPC.
• Experience designing experiments and analyzing results.
• Publications or patents in top-tier venues.

Responsibilities

• Invent new approaches to contextual retrieval, relevance scoring, and LLM-based summarization.
• Fine-tune and evaluate language models for domain-specific understanding.
• Design experiments measuring real customer impact, not just benchmarks.
• Ship production systems and iterate based on live metrics.
• Collaborate with Alexa, Search, and Recommendations as a cross-platform service.
• Mentor team members and shape the technical direction of the roadmap.

Skills

Java

C++

Python

Algorithms & data structures

Statistical analysis

Experiment design

Patents or publications

High-performance computing

Education

PhD or Master's + 4+ years CS/CE/ML

Tools

Unix/Linux

Job description

Amazon is hiring an Applied Scientist to push the boundaries of LLMs, semantic retrieval, and customer understanding at scale. You will analyze large-scale behavioral data, design experiments, and ship production systems with millisecond latency, impacting customer experiences across Amazon’s surfaces.

You’ll work with cross-team partners (Alexa, Search, Recommendations) to define the technical roadmap, mentor others, and translate research into reliable, low-latency solutions that power real