DESCRIPTION

Fulfillment by Amazon (FBA) enables sellers to scale their businesses globally by leveraging Amazon’s world-class fulfillment network. Sellers using FBA benefit from fast, reliable shipping, Prime delivery eligibility, and hassle-free returns—allowing them to focus on growth while we handle operations. The WW FBA Central Analytics team builds and operates scalable, enterprise-grade data infrastructure, tools, and analytics solutions that power WW FBA business. We partner across global product, program, and operations teams to unify diverse datasets, deliver self-service analytics, and develop next-generation capabilities using LLMs to unlock insights.

Our charter includes building the foundational pipelines, governance frameworks, and intelligent interfaces that enable internal customers to query, analyze, and act on complex datasets with natural language. This is an opportunity to work on one of the largest, complex, and critical analytics ecosystems, designing solutions that combine massive scale, high reliability, and advanced AI.

We are seeking a Data Engineer I will support the GenAI-powered insights assistant by building pipelines that process unstructured data (knowledge articles and documents) in the S3 Data Lakehouse. You'll manage vector databases that store embeddings, helping the AI retrieve relevant info quickly and accurately.

Key job responsibilities
- Develop metadata pipelines to tag documents with freshness, ownership, and other context for better filtering.
- Implement caching and multi-region replication to reduce query latency.
- Monitor data retrieval accuracy and log source citations to improve AI trustworthiness.
- Automate ingestion and embedding generation for unstructured data into vector databases like Zilliz, Pinecone, or OpenSearch.