Senior Data Engineer – Retrieval, RAG Platform

This is a fully remote position, open to applicants in United States.

📋 Description

• Establish the Bedrock Knowledge Base and OpenSearch Serverless vector collection within the client’s account utilizing infrastructure code.

• Set up the embedding model, chunking strategy, and metadata fields.

• Execute hybrid vector and keyword retrieval with filtering based on taxonomy, source, status, and review-date metadata.

• Optimize top-k, score thresholds, and reranking in alignment with the golden question suite.

• Develop a versioned retrieval API utilizing API Gateway and Lambda.

• Implement authentication, client-specific rate limits, usage metrics, structured error handling, and citation payloads.

• Enforce entitlement filtering during query execution to ensure restricted passages are not returned to unauthorized users.

• Provide the Bedrock retrieve-and-generate reference endpoint for pilot validation.

• Create a quality dashboard that encompasses relevance, grounding, citation scores, permission tests, freshness lag, API latency and errors, as well as OpenSearch capacity.

• Assist the contact center pilot by identifying retrieval misses, making adjustments, and rerunning the evaluation suite.

• Document retrieval and API runbooks detailing model changes, re-embedding, index rebuilds, and cost implications.

• Collaborate with client platform owners and complete the transition readiness checklist.

⛳️ Requirements

• A minimum of 4 years of experience as a software or ML engineer.

• Proven experience delivering at least one RAG or enterprise search system to production, assessed with an evaluation suite.

• Hands-on expertise with Amazon Bedrock Knowledge Bases, Retrieve and RetrieveAndGenerate APIs, as well as embedding models.

• Familiarity with Amazon OpenSearch Serverless vector search, hybrid queries, and metadata filters.

• Proficiency in Python.

• Experience in deploying Lambda-based APIs behind API Gateway, including aspects such as authentication, versioning, and throttling.

• Knowledge of infrastructure-as-code practices using Terraform or AWS CDK.

• Continuous Integration (CI) experience for deployments into a client-owned account.

• Solid understanding of chunking, embedding model selection, reranking, and the trade-offs related to relevance, latency, and cost.

• Experience in building operational dashboards using CloudWatch or a similar tool.

• Proficiency in English as specified in the application form.

• Legal authorization to work in the country where the position is located, as required by the application form.

🏝️ Benefits

• Equal opportunities for collaboration across diverse continents and cultures.

• An inclusive environment that emphasizes continuous learning, innovation, and adherence to ethical AI standards.