RAG / Search Jobs

A RAG (Retrieval-Augmented Generation) or Search engineer builds the retrieval layer that grounds language models in private or up-to-date data: chunking strategies, embedding models, vector indices, hybrid lexical+semantic ranking, and the LLM-side prompting that turns retrieved passages into trustworthy answers. The role is increasingly its own specialty as enterprise AI deployments hit the limits of naive RAG and need re-rankers, query rewriting, and evaluation harnesses. Strong candidates know information-retrieval fundamentals (BM25, Reciprocal Rank Fusion) and have shipped a production RAG system end-to-end.

As of October 2026, AgenticCareers tracks 63 open RAG / Search positions across 32 companies. The median advertised salary is $190K–$252K per year (based on 21 listings with disclosed pay). 10% of roles are remote. The most active employers are Databricks, MARLOWE COMPANIES INC., Capital One.

Full RAG / Search salary data