This job is no longer accepting applications.Browse similar jobs
Fuzu Atlas
Computers + 1 more
Description
About Fuzu
Fuzu Global Workforce is a service line of Fuzu Oy, headquartered in Helsinki, Finland. We deliver global out-staffing services to forward-thinking businesses, connecting them with top-tier talent across data sciences, languages, technology, data annotation, and creative sectors. With strong expertise in global talent and robust support systems, Fuzu ensures seamless remote collaboration, compliance, and scalable delivery. Our fully vetted consultants are managed by Fuzu, providing high-quality work and the flexibility of global remote teams. Join us in reshaping the future of work—where borders no longer limit opportunity or excellence.
Position Overview
We are seeking creative, analytical, and detail-oriented contractors to help shape the frontier of multimodal AI evaluation. In this role, you will craft complex question-and-answer prompts based on video content to challenge and “stump” large language models (LLMs).
Ideal candidates combine exceptional critical thinking, linguistic precision, and creative problem-solving. Prior experience in adversarial prompt engineering or model evaluation is a strong plus.
This is a remote, contract-based position offering flexible hours and the opportunity to work on cutting-edge AI alignment and evaluation initiatives
Design Complex Prompts: Watch video clips and craft nuanced, context-rich questions and answers designed to challenge advanced LLMs’ reasoning and comprehension.
Analyze Model Behavior: Evaluate how models interpret multimodal cues (audio,visual, and textual) and identify reasoning gaps or inconsistencies.
Synthesize Multimodal Information: Extract key insights, actions, and relationships from video scenes and convert them into structured question–answer pairs.
Iterate and Calibrate: Refine prompts through multiple rounds of testing, incorporating feedback from reviewers and model outputs.
Maintain Consistency and Accuracy: Ensure all content meets project standards for clarity, factual accuracy, logical soundness and more
Desired Background
We welcome applicants from diverse academic and professional backgrounds who demonstrate both creativity and rigor. Candidates may have experience in one or more of the following areas:
Adversarial Prompt Engineering or LLM Evaluation
Creative or Technical Writing, Linguistics, or Journalism
AI Research, Cognitive Science, or Computer Science
Video/Media Analysis or Storytelling
Research & Critical Thinking Roles where deep synthesis and insight generation are key.
Role Requirements
● Advanced degree (Master’s or PhD) in Literature, Linguistics, or related field.
● Proven ability to craft complex, high-quality written content with logical structure and creativity.
● Strong attention-to-detail skills.
● Ability to synthesize visual and audio information into coherent narratives or prompts.
● Familiarity with AI systems, prompt engineering, or LLM evaluation is highly desirable.
● Native English speaker.
● Commitment to confidentiality, accuracy, and consistent quality.
Start hiring with Fuzu
Recruit better talent faster - on your own or with our support.
Explore recruitment platformJob search tips from Fuzu
Selected articles on cover letters, CV structure, and interview preparation.
