Background

DDD believes talent has no boundaries--and opportunities shouldn’t either. In 2001, we saw the need to bring tech skills and living-wage work to men and women in underserved communities in Asia. It was here that DDD helped plant the seed for a socially responsible outsourcing practice known as impact sourcing.

Job Description

We are seeking a highly analytical and detail-oriented Associate – VLA Reviewer to support advanced data review workflows for a leading autonomous vehicle project. This role is focused on reviewing complex Vision-Language Action annotation outputs, ensuring high-quality interpretation of dynamic scenes, agent behavior, object interactions, spatial relationships, and action sequences. The successful candidate will bring hands-on experience in AV data annotation, VLA-related projects, and production-level review workflows. This is an excellent opportunity for a candidate who thrives in quality-focused AI data environments and is comfortable working with complex visual and language-based annotation tasks.

Key Responsibilities

As an Associate – VLA Reviewer, you will be responsible for:

• Reviewing autonomous vehicle data annotation outputs to ensure quality, consistency, and alignment with project guidelines.
• Interpreting complex driving scenes involving multiple agents, objects, interactions, and environmental context.
• Assessing agent actions, intent, object relationships, spatial reasoning, and task sequences.
• Supporting Vision-Language Action workflows, including vision-language alignment, action grounding, temporal understanding, and contextual scene interpretation.
• Identifying and resolving ambiguous or edge-case annotation scenarios using sound judgment and guideline interpretation.
• Applying annotation standards consistently across production review tasks.
• Participating in calibration sessions, QA discussions, and feedback loops to support ongoing quality improvement.
• Providing clear, structured, and actionable feedback to annotators and project stakeholders.
• Escalating unclear guidelines, tooling issues, or recurring quality gaps to project leads as appropriate.

Qualifications

Education Requirements

• Diploma or higher qualification in a relevant field such as: Computer Science, Information Technology, Engineering (Electrical, Computer, Geospatial, or related), Data Science, Geospatial Studies, Or equivalent technical discipline

Required Experience

• Minimum 3 years of experience in autonomous vehicle data annotation.
• Minimum 1 year of experience working on Vision-Language Action projects.
• Experience working with complex scene understanding tasks, including: Object interactions, Agent behavior, Spatial reasoning, Action prediction, Intent interpretation, Task-sequence analysis.
• Experience in review workflows, annotation guideline interpretation, and edge-case handling within production environments.

Required Skills

• Strong understanding of VLA concepts, including vision-language alignment, action grounding, temporal understanding, and contextual scene interpretation.
• Ability to analyze dynamic environments and accurately label or review agent actions, intent, object relationships, and task sequences.
• Excellent attention to detail and strong decision-making skills in ambiguous annotation scenarios.
• Ability to quickly learn project-specific guidelines, tools, workflows, and quality standards with minimal supervision.
• Strong communication and collaboration skills to contribute effectively to calibration sessions, QA discussions, and reviewer feedback loops.
• A disciplined and quality-driven approach to data review and annotation accuracy.

Candidate Profile

The successful candidate will be structured, observant, and comfortable working with complex visual data. They will be able to balance accuracy with productivity, apply detailed guidelines consistently, and contribute to high-quality AI training data for autonomous vehicle systems.

• Reviewing autonomous vehicle data annotation outputs to ensure quality, consistency, and alignment with project guidelines.
• Interpreting complex driving scenes involving multiple agents, objects, interactions, and environmental context.
• Assessing agent actions, intent, object relationships, spatial reasoning, and task sequences.
• Supporting Vision-Language Action workflows, including vision-language alignment, action grounding, temporal understanding, and contextual scene interpretation.
• Identifying and resolving ambiguous or edge-case annotation scenarios using sound judgment and guideline interpretation.
• Applying annotation standards consistently across production review tasks.
• Participating in calibration sessions, QA discussions, and feedback loops to support ongoing quality improvement.
• Providing clear, structured, and actionable feedback to annotators and project stakeholders.
• Escalating unclear guidelines, tooling issues, or recurring quality gaps to project leads as appropriate.

• Strong understanding of VLA concepts, including vision-language alignment, action grounding, temporal understanding, and contextual scene interpretation.
• Ability to analyze dynamic environments and accurately label or review agent actions, intent, object relationships, and task sequences.
• Excellent attention to detail and strong decision-making skills in ambiguous annotation scenarios.
• Ability to quickly learn project-specific guidelines, tools, workflows, and quality standards with minimal supervision.
• Strong communication and collaboration skills to contribute effectively to calibration sessions, QA discussions, and reviewer feedback loops.
• A disciplined and quality-driven approach to data review and annotation accuracy.

• Diploma or higher qualification in a relevant field such as: Computer Science, Information Technology, Engineering (Electrical, Computer, Geospatial, or related), Data Science, Geospatial Studies, Or equivalent technical discipline
• Minimum 3 years of experience in autonomous vehicle data annotation.
• Minimum 1 year of experience working on Vision-Language Action projects.
• Experience working with complex scene understanding tasks, including: Object interactions, Agent behavior, Spatial reasoning, Action prediction, Intent interpretation, Task-sequence analysis.
• Experience in review workflows, annotation guideline interpretation, and edge-case handling within production environments.