Applied Scientist: Fraud Detection with ML/NLP

Amazon Inc.; Omaha (NE); On-site; USD 172,000 - 223,000; Full time

Benefits offered by this job

Health insurance

401(k) matching

Paid time off

Parental leave

RSUs

Job summary

Audible is seeking an Applied Scientist to build fraud-detection systems using ML, DL, and NLP. You will scale data pipelines, design predictive models, and protect customers and creators from AI-generated fraud in a fast-paced, agile environment.

You’ll collaborate with data scientists, ML engineers, and business partners worldwide, publish ideas and artifacts, and contribute to enterprise risk management while advancing state-of-the-art fraud defenses.

Qualifications

• MS in CS or related field with 5+ years of experience, or PhD
• Fluent in SQL, Python
• Depth and breadth in novel machine learning technologies
• Machine Learning Pipeline orchestration with AWS (SageMaker, Batch, Lambda, Step Functions) or similar cloud-platforms
• Big Data Engineering with Spark/AWS EMR & Glue
• Experience with Agile Software Development

Responsibilities

• Protect Audible’s customers and content creators against the onslaught of AI-generated fraud
• Develop Amazon-scale data engineering & modelling pipelines
• Invent fraud detection and mitigation solutions using innovative approaches
• Collaborate with data scientists, ML engineers, engineers and business teams across the globe
• Contribute to the growth of the Audible Data Science team by sharing ideas and learnings

Skills

SQL

Python

Agile Software Development

Machine Learning

AWS SageMaker

Spark

EMR/Glue

Education

MS in Computer Science

PhD

Tools

SageMaker

Batch

Lambda

Step Functions

Spark

EMR

Glue

Job description

Audible is seeking an Applied Scientist to build fraud-detection systems using ML, DL, and NLP. You will scale data pipelines, design predictive models, and protect customers and creators from AI-generated fraud in a fast-paced, agile environment.

You’ll collaborate with data scientists, ML engineers, and business partners worldwide, publish ideas and artifacts, and contribute to enterprise risk management while advancing state-of-the-art fraud defenses.