Applied Scientist (L4), AEE Science and AI

AEE Science and AI team is focused on advancing customer experience on 1) Expansions: building the end-to-end Day0 search relevance models (ranking and matching) for new Amazon marketplaces, 2) Exports: improving the search relevance for cross-border (exports) customers, and 3) Emerging Markets Retail Efficiency related initiatives. We focus on algorithms and machine learning model developments, and optimize for recall (completeness) and precision (accuracy), measured through Spare Result Rate (SRR) and Irrelevant Result Rate (IRR) respectively. We are also leveraging Generative AI and LLMs to improve retail operations and reduce cost-to-serve.

In this role, you will leverage your expertise in machine learning, information retrieval, and natural language processing to design, implement, and optimize ranking and matching algorithms that improve user experience and satisfaction. Key responsibilities include analyzing large datasets to identify patterns and insights, developing and fine-tuning models to improve search result relevance, and collaborating with cross-functional teams to integrate these models into our search systems. You will conduct experiments to evaluate model performance, utilizing A/B testing and other methodologies to ensure continuous improvement.

The ideal candidate will possess a strong background in applied mathematics, computer science, or a related field, with experience in search technologies and a passion for solving complex problems. You will have the opportunity to make a significant impact on our product by driving advancements in search relevance that directly enhance user engagement and outcomes.

Key job responsibilities
1. Model Development: Design, implement, and refine machine learning models for ranking and matching to enhance search relevance and user satisfaction.

2. Data Analysis: Analyze large datasets to extract insights, identify trends, and inform model improvements, ensuring data-driven decision-making.

3. Algorithm Optimization: Continuously evaluate and optimize existing algorithms to improve search accuracy and performance through experimentation and iteration.

4. Performance Evaluation: Conduct A/B testing and other evaluation methodologies to assess model effectiveness, drawing actionable conclusions to guide further development.

5. Cross-Functional Collaboration: Work closely with product managers, engineers, and UX researchers to integrate search models into production systems and align with overall product goals.

6. Research and Innovation: Stay current with advancements in machine learning, natural language processing, and information retrieval to drive innovation in search technologies.

7. Reporting: Document methodologies, findings, and model performance metrics, and communicate results effectively to stakeholders at all levels.