Staff Video Data & Model Evaluation Engineer

Cantina; Singapore; On-site; SGD 120,000 - 180,000; Full time

Benefits offered by this job

Competitive salary and generous equity

Personal time off and paid holidays

Health insurance

Global travel insurance

Monthly spending stipend: $500 (~S$635

Equipment: All equipment needed for my

Job summary

Cantina is seeking a Member of Technical Staff to join our Singapore research team. You will shape the data and evaluation strategies behind our video generation models, owning the full data lifecycle from requirements to high‑quality datasets and evaluation feedback.

You will collaborate with research scientists, ML engineers, product teams, and external data partners to translate analysis into model improvements and scalable data workflows for robust, real‑time generation.

Qualifications

• Define data requirements and annotation standards for video generation datasets.
• Develop labeling taxonomies and captioning guidelines for quality and coverage.
• Establish evaluation frameworks to measure model outputs and failure modes.
• Coordinate with researchers and engineers to align data strategy with model goals.
• Implement automated data pipelines and quality controls for large-scale datasets.

Responsibilities

• Define quality standards for video generation across aesthetics, realism, composition, motion, and text rendering.
• Evaluate model iterations, analyze failure patterns, and translate findings into actionable data and prompting strategies.
• Own end-to-end training data projects, including sourcing, construction, annotation, captioning, QC, and delivery.
• Develop labeling systems for dataset segmentation, coverage analysis, quality filtering, and model attribution.
• Design annotation and captioning standards for video generation, editing, and multi-reference tasks.
• Create and evaluate synthetic-data workflows based on stability, usability, quality, and training value.
• Develop automated annotation and filtering workflows using vision-language models and prompts.
• Build structured model evaluation processes covering test-set design and reporting.
• Establish feedback loops connecting model evaluation with data strategy and model iteration.
• Train annotation teams and external partners, conduct quality audits, and improve delivery standards.
• Partner with research, engineering, product, and creative teams on data initiatives.

Skills

Data strategy

Dataset development

Annotation

Model evaluation

Generative media

Vision-language models

Prompt design

Python

SQL

Tools

Python

SQL

Job description

Cantina is seeking a Member of Technical Staff to join our Singapore research team. You will shape the data and evaluation strategies behind our video generation models, owning the full data lifecycle from requirements to high‑quality datasets and evaluation feedback.

You will collaborate with research scientists, ML engineers, product teams, and external data partners to translate analysis into model improvements and scalable data workflows for robust, real‑time generation.