true
Mid-level
<p>Please submit your resume in English and indicate your level of English proficiency.</p><p>Mindrift connects specialists with project-based AI opportunities for leading tech companies, focused on testing, evaluating, and improving AI systems. Participation is project-based, not permanent employment.</p><p>What this opportunity involves </p><p>This project requires you to put yourself in the position of a range of different user personas and engage in realistic multi-turn conversations with LLMs, working towards a clearly defined goal. You will need to: </p><ul><li>use a range of tones and registers, </li><li>stress-test the models’ ability to respond adequately based on several abstract dimensions (e.g. instruction-following, emotional intelligence, consistency under changing constraints), and </li><li>react and adapt to model output while maintaining tight focus on each individual task’s requirements. </li></ul><p>Think of yourself as a controlled adversary, crafting plausible human dialogue that exposes subtle model weaknesses while maintaining narrative coherence.</p><p>For this, you will need to:</p><ul><li>Think like a storyteller and a tester</li><li>Understand how humans actually speak, hesitate, contradict themselves, and escalate emotionally</li><li>Be able to deliberately engineer conversational pressure without breaking realism</li><li>Be methodical enough to document observations clearly and consistently</li><li>Be able to pinpoint failure modes and LLM patterns</li></ul><p>What we look for</p><p>This opportunity is a good fit if you are seeking for open to part-time, non-permanent projects. Ideally, contributors will have: </p><ul><li>an under- or postgraduate qualification in an Arts-based subject (English, Creative Writing, Journalism, MFL, Psychology, Cognitive Science), related field, or work experience at an equivalent level; or 1+ years’ experience in Conversational AI Testing, Narrative Design, or Adversarial Model Testing</li><li>C2-level English (CPE, TOEFL 114+, IELTS 8.0 or above) </li></ul><p>Nice to have</p><ul><li>Conversational UX / dialogue design experience</li><li>An understanding of prompt engineering or LLM evaluation</li><li>Experience with QA testing for complex systems</li><li>A background in narrative design, interactive fiction, or screenwriting</li><li>A qualification in, or professional experience with, behavioural research, psychology, or linguistics</li><li>Demonstrated familiarity with LLM behaviour, failure modes, and evaluation concepts</li><li>Experience working with structured guidelines, rubrics, or annotation frameworks</li></ul><p>You will receive training in our guidelines and in how to create structured, focused conversations that meet the project’s goals. You will also be assigned a mentor who will guide you through your first conversations and provide clear, actionable feedback to support your improvement.</p><p>How it works </p><p>Apply → Pass qualification(s) → Join a project → Complete tasks → Get paid </p><p>Project time expectations </p><p>For this project, tasks are estimated to require around 10–20 hours per week during active phases, based on project requirements. This is an estimate, not a guaranteed workload, and applies only while the project is active. </p><p>Payment </p><ul><li>Paid contributions, with rates up to $25/hour* </li><li>Fixed project rate or individual rates, depending on the project </li><li>Some projects include incentive payments </li></ul><p>*Note: Rates vary based on expertise, skills assessment, location, project needs, and other factors. Higher rates may be offered to highly specialized experts. Lower rates may apply during onboarding or non-core project phases. Payment details are shared per project.</p>