Job Description

Team Introduction

Safety Model Operations (SMO) is a dedicated organization that provides end-to-end support for AI Moderation services across a wide range of international products. Our core function is to deliver high-quality labeling, evaluation, and safety-related data services that help ensure AI systems operate responsibly, accurately, and in alignment with product requirements.

The SMO Delivery Team plays a critical role in the organization. Its primary responsibility is to carry out the full spectrum of quality assurance activities for each project. This includes:

- Conducting detailed reviews and complex RCA's to ensure labeling accuracy and consistency

- Monitoring quality performance and compliance against project-specific KPIs

- Identifying trends, risks, and potential gaps in processes or guidelines

- Providing structured feedback and improvement recommendations to the Central Project Team

- Supporting continual optimization of workflows, tools, and evaluation methodologies

- Improve Model performance of AI models

What will I do?

As Group Leader, you will lead and scale large operational teams within Safety Model Operations across APAC. You will drive operational excellence while ensuring teams deeply understand the link between high-quality human operations (content moderation, data labeling, red-teaming, and evaluation) and improved AI model performance. This role involves coaching team leads, managing cross-functional stakeholders, and actively collaborating with product, engineering, and policy teams to optimize annotation workflows and close the loop between operations and model outcomes.

Responsibilities

- Lead, motivate, and develop large, fast-paced operational teams to deliver high-quality AI moderation and labeling services, with clear accountability for both operational KPIs and downstream model impact.

- Champion a strong understanding of how operations (human content moderation, annotation quality, guideline design, and rater performance) affect model training, safety, and alignment — including RLHF/RLAIF feedback loops, bias introduction, and safety regressions.

- Conduct detailed reviews, complex root cause analyses (RCAs), and quality audits while linking findings to model-level outcomes (e.g., safety benchmark improvements, reduced hallucinations, or lower post-deployment incidents).

- Monitor quality performance and compliance against project KPIs; proactively identify trends, risks, and gaps in processes, guidelines, or rater pools that could degrade model performance.

- Drive continuous improvement initiatives focused on annotation accuracy, inter-annotator agreement, cultural/contextual sensitivity, and the translation of human feedback into better model generalization.

- Collaborate cross-functionally with product, engineering, policy, and data science teams to run experiments, measure the impact of operational changes on model metrics, and enhance proprietary tools, systems, and workflows.

- Build and maintain training programs that equip teams with both operational excellence and a systems-thinking mindset on data-model feedback loops.

- Manage data collection, reporting, productivity, unit costs, and quality targets while ensuring exceptional delivery and regulatory compliance.

- Identify, escalate, and resolve operational issues, removing barriers and driving alignment across stakeholders.