Member of Technical Staff, Video Data & Model Evaluation
Core
Define data strategies, build high-quality video generation datasets, and establish evaluation frameworks to improve generative video models.
Role type
Senior IC data strategy and model evaluation engineer (video generation)
Builds
High-quality training datasets, evaluation pipelines, and automated annotation workflows for video generation models
Domain
Generative AI, video production, multimodal models
Deliverable
production ML models
Required skills
Data strategy, dataset development, annotation standards, model evaluation, visual judgment, project management, quality control
Preferred skills
Synthetic data workflows, vision-language models, prompt design, Python, SQL
Technologies
Vision-language models, system prompts, quality-scoring models, Python, SQL
Responsibilities
Define quality standards for video generation aesthetics and realism; Evaluate model iterations and analyze failure patterns; Own end-to-end training data projects including sourcing and annotation; Develop labeling systems and annotation standards; Create and evaluate synthetic-data workflows; Build structured model evaluation processes; Establish feedback loops between evaluation and data strategy; Train and calibrate annotation teams and external partners
Seniority
Senior, hands-on IC
