Search Relevance Analyst

Globik AI; United States; Remote; USD 14,000 - 22,000; Full time

Job summary

Globik AI is seeking Search Query–Result Relevance Specialists to evaluate whether search results genuinely match what users are looking for.

The role focuses on reviewing queries alongside returned results, judging relevance and usefulness, and providing consistent ratings to improve search systems. You will participate in calibration, training, and quality-review sessions to maintain high accuracy across large evaluation tasks.

Qualifications

• Strong reading comprehension and analytical judgment.
• Excellent attention to detail.
• Experience in search evaluation content review QA research or AI evaluation is a plus.
• Ability to understand search queries and determine user intent.

Responsibilities

• Review user search queries and returned search results.
• Determine whether each result directly addresses the user's intent.
• Rate results according to relevance, usefulness, freshness, and project criteria.
• Identify unrelated, partially relevant, misleading, duplicate, or low-quality results.
• Distinguish genuinely useful results from results that only share keywords with the query.
• Consider ambiguity and likely user expectations when making judgments.
• Apply project-specific relevance guidelines consistently.
• Document difficult or ambiguous cases and elevate them when required.
• Maintain accuracy across large evaluation batches.
• Participate in calibration and quality-review sessions.

Skills

Reading comprehension

Analytical judgment

Attention to detail

Understand user intent

Quality mindset

Job description

Globik AI is seeking Search Query–Result Relevance Specialists to evaluate whether search results genuinely match what users are looking for.

The role focuses on reviewing queries alongside returned results, judging relevance and usefulness, and providing consistent ratings to improve search systems. You will participate in calibration, training, and quality-review sessions to maintain high accuracy across large evaluation tasks.