RAG Data Engineer

Booz Allen Hamilton•Annapolis Junction, MD

7h•$61,900 - $141,000•Onsite

About The Position

As a RAG Data Engineer, you will build and operate the data ingestion and search infrastructure that powers mission-critical Retrieval-Augmented Generation (RAG) capabilities. You will design, implement, and maintain pipelines that transform structured and unstructured information into retrieval-ready assets for AI-powered applications. You will own ingestion, embedding, and indexing workflows connecting enterprise data sources with Amazon Bedrock and OpenSearch. This includes document normalization, chunking, embedding generation, metadata enrichment, lineage tracking, and reliable upsert or delete operations. You will also design and optimize OpenSearch indexes, mappings, shard configurations, lifecycle policies, and query performance. Working with cloud architects, security engineers, application developers, and AI/ML specialists, you will help establish a scalable, secure, and resilient data retrieval foundation supporting enterprise RAG application development and deployment. The position is located in Annapolis Junction, Maryland, with Fort Meade, Maryland, listed as an additional location; remote work is not offered. TS/SCI clearance with a polygraph is required.

Requirements

Experience developing and maintaining ETL or data ingestion pipelines for structured and unstructured data.
Experience implementing document normalization, parsing, chunking, and content preparation workflows; generating embeddings; and integrating embedding workflows into data processing pipelines.
Experience designing and maintaining OpenSearch or Elasticsearch indexes, schemas, mappings, and aliases, and optimizing search performance, including query tuning, indexing efficiency, and shard management.
Experience implementing idempotent ingestion processes, including upsert, delete, replay, and failure recovery.
Experience with metadata modeling, enrichment, lineage, and version tracking.
Experience with Python or other programming languages for data engineering and automation.
Experience with containerized applications or deployment environments such as Docker, ECS, EKS, or Kubernetes.
Ability to collaborate with engineering, security, cloud, and application development teams.
TS/SCI clearance with a polygraph.
High school diploma or GED.

Nice To Haves

Experience integrating Amazon Bedrock embeddings into RAG ingestion pipelines.
Experience tuning hybrid retrieval systems combining BM25 keyword search and vector similarity search.
Experience implementing OpenSearch lifecycle policies, snapshots, restoration, and index migration strategies.
Experience with distributed or asynchronous data processing using Spark, Ray, Dask, Kafka, or Amazon SQS.
Experience optimizing document chunking, embeddings, and indexing strategies using retrieval evaluation results.
Experience handling multi-source data ingestion and cross-source normalization.
Experience designing scalable, secure data infrastructure using AWS services.
Experience with infrastructure-as-code tools such as Terraform, CloudFormation, or AWS CDK.

Responsibilities

Build, operate, design, implement, and maintain data ingestion pipelines for structured and unstructured information.
Own ingestion, embedding, and indexing workflows that connect enterprise data sources with Amazon Bedrock and OpenSearch.
Implement document normalization, chunking, embedding generation, metadata enrichment, lineage tracking, and reliable upsert or delete operations.
Design and optimize OpenSearch indexes, mappings, shard configurations, lifecycle policies, and query performance.
Collaborate with cloud architects, security engineers, application developers, and AI/ML specialists to establish a scalable, secure, and resilient data retrieval foundation.
Support the development and deployment of enterprise RAG applications.