Evaluation Scenario Writer - AI Agent Testing Specialist

2 days ago Be among the first 25 applicantsThis opportunity is only for candidates currently residing in the specified country. Your location may affect eligibility and rates. Please submit your resume in English and indicate your level of English.At Mindrift, innovation meets opportunity. We believe in using the power of collective human intelligence to ethically shape the future of AI.What We DoThe Mindrift platform connects specialists with AI projects from major tech innovators. Our mission is to unlock the potential of Generative AI by tapping into real-world expertise from across the globe.About The RoleWe''re looking for someone who can design realistic and structured evaluation scenarios for LLM-based agents. You''ll create test cases that simulate humanperformed tasks and define goldstandard behavior to compare agent actions against. You''ll work to ensure each scenario is clearly defined, wellscored, and easy to execute and reuse. You''ll need a sharp analytical mindset, attention to detail, and an interest in how AI agents make decisions.ResponsibilitiesCreate structured test cases that simulate complex human workflowsDefine goldstandard behavior and scoring logic to evaluate agent actionsAnalyze agent logs, failure modes, and decision pathsWork with code repositories and test frameworks to validate your scenariosIterate on prompts, instructions, and test cases to improve clarity and difficultyEnsure that scenarios are productionready, easy to run, and reusableRequirementsBachelor''s and/or Master''s Degree in Computer Science, Software Engineering, Data Science / Data Analytics, Artificial Intelligence / Machine Learning, Computational Linguistics / Natural Language Processing (NLP), Information Systems or other related fields.Background in QA, software testing, data analysis, or NLP annotationGood understanding of test design principles (e.g., reproducibility, coverage, edge cases)Strong written communication skills in EnglishComfortable with structured formats like JSON/YAML for scenario descriptionCan define expected agent behaviors (gold paths) and scoring logicBasic experience with Python and JSCurious and open to working with AIgenerated content, agent logs, and promptbased behaviorNice to HaveExperience in writing manual or automated test casesFamiliarity with LLM capabilities and typical failure modesUnderstanding of scoring metrics (precision, recall, coverage, reward functions)BenefitsGet paid for your expertise, with rates that can go up to $45/hour depending on your skills, experience, and project needsTake part in a flexible, remote, freelance project that fits around your primary professional or academic commitmentsParticipate in an advanced AI project and gain valuable experience to enhance your portfolioInfluence how future AI models understand and communicate in your field of expertiseSeniority LevelEntry levelEmployment TypeParttimeJob FunctionOtherIndustriesIT Services and IT Consulting #J-18808-Ljbffr

Created: 2026-04-25