AI Architect - RAG & Data Architecture

Design scalable architectures that ingest, process, retrieve and reason over large volumes of structured and unstructured enterprise data, with deep expertise in retrieval-augmented generation, embeddings, vector databases and enterprise data pipelines.

Responsibilities:

• Architect enterprise RAG solutions for structured and unstructured data.
• Design scalable ingestion, ETL and retrieval architectures for large enterprise datasets.
• Define embedding strategies for documents and other enterprise information.
• Design storage and retrieval architectures using SQL, NoSQL and vector databases.
• Define chunking, indexing, retrieval, ranking and grounding strategies.
• Architect solutions using platforms such as Elastic or Pinecone.
• Establish data-quality and AI model-governance practices.
• Design methods to reduce unsupported LLM responses.
• Define cloud-native AI and data architecture.
• Guide engineering teams on implementation standards and architectural practices.

Requirements:

• 10+ years of overall professional experience, including 3+ years of directly relevant hands-on experience.
• Deep understanding of RAG architecture, vector embeddings and vector databases.
• Experience processing structured and unstructured data with SQL and NoSQL technologies.
• Strong Python skills and experience with PySpark or other big-data technologies.
• Experience with AWS, Azure or Google Cloud.
• Experience with Databricks and/or Snowflake and ETL or data-pipeline architecture.
• Understanding of LLMs, Generative AI, prompt engineering and model fine-tuning concepts.

Preferred qualifications:

• Experience designing enterprise-scale AI architecture.
• Experience with data governance.
• Understanding of AI security and compliance requirements.
• Experience with large-scale document ingestion and retrieval systems.