Search & Retrieval Engineer

Role overview Improve how AI agents find and use information in business documents such as policies, prices, FAQs, and other materials that may change over time. You’ll prepare and index knowledge, tune retrieval, and evaluate whether changes help agents answer accurately, handle conflicting versions, and acknowledge when information is missing.

Responsibilities - Improve document preparation, splitting, indexing, and retrieval. - Address outdated or conflicting material, including multiple versions of a policy. - Build evaluation sets with expected answers and supporting sources. - Compare answer quality before and after retrieval changes using clear measures. - Ensure agents can acknowledge when the available knowledge does not contain an answer.

Requirements - Experience with search, information retrieval, or retrieval-augmented generation. - Python skills and practical knowledge of embeddings, vector search, and ranking. - Rigor in evaluation, including fixed test sets and clear metrics. - Ability to explain changes and results clearly in writing. - Relevant experience may include search or ML engineering, support or documentation retrieval systems, NLP, or data engineering for knowledge products.

Benefits and work setup Remote contract with flexible hours. Compensation starts at $5,500 per month. Freelancers and solo agencies may continue working with their own clients. The selection exercise uses a small document set with conflicting versions: prepare control questions, improve data preparation or search, then compare answers and sources before and after the change. There is no time limit.