Please submit your resume in English and indicate your level of English. At Toloka, we connect smart, curious people from around the world with freelance online tasks that train and improve artificial intelligence. What we do The Toloka Annotators connects individuals with Generative AI projects from leading tech innovators. Our mission is to unlock the full potential of AI by involving real people from around the world in the development process. About the Role Annotation is what helps AI make sense of the world. As an annotator, you may be invited to take part in online projects such as rating AI-generated content, evaluating factual accuracy, or comparing responses - when projects are available. Responsibilities: - Carefully review provided data (text, images, or videos) - Label or classify content based on project guidelines - Identify and flag factually incorrect, sensitive, inappropriate, or unclear material Important note: This is project-based work. Tasks are available only when projects are active. You may be invited to one or more projects depending on your profile and current opportunities. Each project has its own compensation level based on scope and expertise required. On this project, AI trainers earn up to $18 per hour equivalent.

• Design and implement machine learning algorithms to analyze brain-related data and generate predictive insights into cognitive functions. • Collaborate with the research team in planning experiments, data collection, and analysis. • Develop intuitive data visualization tools to communicate complex research outcomes effectively. • Contribute to the creation of AI-driven models for personalized learning and educational assessments. • Share research findings with the team and support the preparation of scientific publications. • Engage in brainstorming discussions, bringing forward innovative ideas that align with the company’s research vision.

• Build and maintain end-to-end data pipelines for large-scale image and video datasets: collection, filtering, augmentation, conditioning alignment, and efficient storage/sampling. • Implement model architectures (diffusion, autoregressive, flow-based, diffusion transformers, etc.) and maintain high-throughput PyTorch training loops for large-scale image and video diffusion models. • Run and manage large-scale training experiments on multi-GPU and multi-node setups (DDP, FSDP, DeepSpeed). Debug training instabilities, loss spikes, and convergence issues. • Apply quantization, pruning, and knowledge distillation techniques to compress models without sacrificing quality. • Collaborate with researchers and translate state-of-the-art research papers into working implementations in our internal codebase (e.g., new attention mechanisms, sampling schedules, or conditioning methods). • Build and maintain evaluation pipelines of image quality, video consistency, and perceptual metrics. • Set up and maintain human annotation and evaluation pipelines using services like AWS GroundTruth. • Profile and optimize training speed, GPU memory utilization, and iteration time. Implement inference optimizations to reduce latency and compute cost. • Work with acceleration toolchains such as torch.compile, Triton, TensorRT, or ONNX where appropriate

Role Description - Write and refine prompts to guide model behavior in physics contexts. - Evaluate LLM-generated responses to physics-related queries for conceptual accuracy and reasoning quality. - Conduct fact-checking using authoritative public sources and domain knowledge. - Annotate model responses by identifying strengths and areas of improvement. - Assess clarity, structure, and appropriateness of explanations for different audiences. - Apply consistent evaluation standards by following clear taxonomies and benchmarks. Qualifications - Must-Have: - PhD in Physics or closely related field. - Deep expertise in Classical Mechanics & Dynamics, Electromagnetism, Thermodynamics, Quantum Physics, or Relativity. - Significant experience using large language models (LLMs). - Excellent writing skills and ability to explain complex physics concepts. - Strong attention to detail and experience reviewing technical writing. - Preferred: - Experience with RLHF, model evaluation, or data annotation work. - Experience teaching or explaining physics concepts to non-experts. - Familiarity with evaluation rubrics or structured review frameworks. Requirements - Full-time or Part-time Contract Work - Compensation: $73/hour - Location: USA, UK, Canada, EU Application Process - Upload resume - AI interview based on your resume - Submit form Resources & Support - For details about the interview process and platform information, please check: Interview Process - For any help or support, reach out to: support@mercor.com - Our team reviews applications daily. Please complete your AI interview and application steps to be considered for this opportunity.