Hiring nowInformation retrieval and grounding

Retrieval-Augmented Generation Engineer

Also called RAG Engineer

A RAG engineer builds the service that finds the right evidence and makes generated answers stick to it. It matters because users trust answers only when they are supported, fast and checkable.

What they do

The work starts with the source material. They decide how documents are cleaned, split and indexed, and keep track of where each passage came from. They tune retrieval with search engines, vector indexes and rerankers, then test whether the model actually uses what it was given. They build evaluation sets that score relevance and answer support, and run them on every change. They read failed answers one by one to find whether retrieval or generation caused the miss. They also watch latency and cost, and trade them against quality.

What a strong hire delivers

In the first months, a strong hire puts a working RAG service into use with clear evaluation behind it. The team can see the supported-answer rate, retrieval recall and latency on one page. Source data has known provenance and a repeatable ingestion path. Failure cases are sorted by cause. Changes to chunking, retrieval or prompts are judged against evidence, not opinion.

How you can tell they are good

Ask them to walk through a RAG system they built and where it failed. A strong candidate separates retrieval errors from generation errors without hesitation. They can explain how they built their test set, what they chose to measure and what the numbers missed. Ask how they would check that an answer is supported by its source. Be wary of anyone who only talks about model choice or prompts.

Why it is hard to fill

The title is new and the skills are split across older ones. Many people have built a demo. Far fewer have measured retrieval quality and fixed it in production. Good candidates may be called search engineer, NLP engineer or applications engineer. Some strong people from information science have never used the RAG label. Hiring on title alone misses them, and hiring on keywords finds plenty of shallow experience.

Where to find them

Look first in search and relevance teams, where people already tune ranking and judge result quality. NLP engineers and information scientists are another strong source. Application engineers who have shipped assistants over private data often fit too. They tend to sit in AI product companies, enterprise software firms, knowledge management teams and search-focused businesses. Ask about evaluation work, not just what they built.

Indicative pay

Not enough published evidence yet to give a fair range for this role, which is common for work this new. We benchmark it for you as part of a search.

Interview questions for a Retrieval-Augmented Generation Engineer

  1. Tell us about a RAG system you shipped. What was wrong with it at first, and how did you find out?

    A strong answer names a specific failure, how it was measured and what changed. Listen for error analysis on real queries. Worry if they describe only the architecture, or say it worked well without evidence.

  2. How do you decide chunk size and splitting strategy for a new document set?

    Good answers start from document structure and query types, then test options against a labelled set. They mention trade-offs in context and recall. Worry about a fixed default applied to everything.

4 more questions, with what a strong answer covers, are in the Role Kit.

Job description you can post

We are building products that answer questions from our own and our customers' content. We need a RAG engineer to own the path from source documents to supported answers. You will work with product and engineering to make answers accurate, fast and easy to check.

What you will do

  • Build and run the retrieval and generation service behind our answers
  • Design ingestion, chunking and indexing for varied source documents
  • Tune retrieval with search engines, vector indexes and rerankers
  • Create evaluation sets for relevance and answer support
  • Analyse failed answers and trace them to retrieval or generation
  • Track supported-answer rate, retrieval recall and latency
  • Record the provenance of source content

What you will bring

  • Shipped a retrieval-based system used by real users
  • Evidence of measuring retrieval quality and acting on the results
  • Practical experience with search engines or vector indexes
  • Strong Python and service engineering skills
  • Clear approach to testing whether answers are grounded in sources
  • Able to explain trade-offs between quality, latency and cost

Nice to have

  • Background in search relevance or information science
  • Experience with rerankers
  • Experience handling messy enterprise documents

Free to use and adapt. If it saves you time, a link back to this page is appreciated.

Skills that matter

  • Information retrievalAdvancedRetrieve and rank relevant documents or entities
  • Retrieval-augmented generationAdvancedGround generated responses in retrieved evidence
  • Data quality and provenanceWorkingTrack dataset origin, validity and transformations
  • Model evaluationWorkingMeasure capability and failure modes credibly

Other roles in information retrieval and grounding

Hiring nowSearch Relevance EngineerA Search Relevance Engineer improves how well a system finds the right evidence for the queries that matter. Better retrieval means grounded…Hiring nowEnterprise Search ArchitectAn Enterprise Search Architect designs search that finds the right content across an organisation while respecting who is allowed to see what. It…EmergingGrounding Evaluation SpecialistA Grounding Evaluation Specialist tests whether an AI system's claims actually follow from the evidence it was given. They make unsupported answers…

Related roles in other families

Hiring nowKnowledge Graph EngineerKnowledge graphs and semantic systemsA Knowledge Graph Engineer builds linked records of the people, products and things a business cares about, each traced to its source. It gives AI…Hiring nowOntology EngineerKnowledge graphs and semantic systemsAn Ontology Engineer defines the concepts, relationships and rules that describe a business domain, so retrieval and AI systems return answers that…Hiring nowAI Knowledge ArchitectKnowledge graphs and semantic systemsAn AI Knowledge Architect designs how an application organises, governs and updates what it knows. That structure decides whether retrieval is…EmergingNeuro-Symbolic Knowledge EngineerKnowledge graphs and semantic systemsA neuro-symbolic knowledge engineer connects structured knowledge, such as ontologies and graphs, to learned models. The aim is answers that stay…

How we wrote this

Role definitions are our own editorial work, checked against occupational classifications, standards, research and employer pages. Titles vary by company, and this is a description of the work, not a job advert.

← Information retrieval and groundingAll new job families

Start a search

Hiring a Retrieval-Augmented Generation Engineer?

Tell us the work. We find the people already doing it, whatever their job title says, and you meet them already interviewed for your role.