Generative Media Quality Engineer

Vamp; Menlo Park (CA); On-site; USD 131,000 - 190,000; Full time

Benefits offered by this job

Bonus

Equity

Benefits

Job summary

Meta's Product Content Engineering team in the US seeks a Content Engineer focused on model training for image and video generation. You will own the human evaluation layer, design rubrics, run evaluation rounds, and translate subjective judgments into measurable signals shaping model behavior and launches.

You will collaborate with product, research and modeling teams, maintain evaluation rubrics, curate test sets, and translate human judgments into automated scores to support scalable,

Qualifications

• 6+ years of experience in video production, content strategy, editorial standards, media evaluation or relevant fields.
• Hands-on video production experience, including shooting, editing, post and/or motion work, with the craft vocabulary to describe quality failures precisely
• Experience creating or substantially contributing to evaluation rubrics or quality guidelines for video or image content
• Experience producing or evaluating short-form video and emerging video formats
• Experience applying editorial judgment, plus the ability to document judgment logic into reusable frameworks that scale across teams
• Experience leveraging quantitative insights to inform content and quality decisions
• Experience collaborating with cross-functional teams (product, engineering, design, research, analytics)
• Bachelor's degree in a directly related field, or equivalent practical experience

Responsibilities

• Build and maintain evaluation rubrics for image and video generation quality, including dimension definitions, rating scales and annotator guidelines
• Execute human evaluation rounds on model output, side-by-sides, absolute scoring and prompt-adherence checks, and produce clear, actionable writeups
• Construct and curate prompt sets and golden sets that stress-test specific model capabilities and failure modes
• Calibrate with fellow raters and vendor teams
• help drive inter-rater reliability
• Partner with research and modeling teams so eval findings translate into training and post-training signal
• Support the transition of human judgment into automated measurement and help validate automated scores against human ground truth
• Track quality trends across model versions and flag regressions early
• Deliver high-quality evaluation work on model and launch timelines in a fast-paced, dynamic environment

Skills

Video production

Editorial standards

Media evaluation

Cross-functional collaboration

Quantitative analysis

Independent working

Education

Bachelor's degree

Tools

Annotation platforms

Eval tooling

Job description

Meta's Product Content Engineering team in the US seeks a Content Engineer focused on model training for image and video generation. You will own the human evaluation layer, design rubrics, run evaluation rounds, and translate subjective judgments into measurable signals shaping model behavior and launches.

You will collaborate with product, research and modeling teams, maintain evaluation rubrics, curate test sets, and translate human judgments into automated scores to support scalable,