AI Model & Agent Ops Intern — Drive AI Quality

TikTok; Singapore; On-site; SGD 13,000 - 23,000; Part time

Job summary

TikTok is offering an AI Model and Agent Operation Project Intern position (Business Integrity) starting in 2026 in Singapore. You will translate moderation policies into machine-executable rules and collaborate across Policy, Product, Engineering, ML, and Operations to drive model improvements.

You will monitor model metrics, perform root-cause analysis on errors, and support evaluation and strategy validation to enhance model quality and consistency.

Qualifications

• Pursuing an Undergraduate or Master’s degree in CS, AI, data science, statistics or a related AI discipline.
• Strong interest in AI, ML, Trust & Safety, or AI Model Operations.
• Strong analytical and problem-solving skills with data-driven insights.
• Familiarity with Excel and SQL; Python or basic ML concepts are a plus.
• Excellent communication and cross-functional collaboration skills.
• Available to intern at least 4 days per week for a minimum of 3 months.

Responsibilities

• Translate moderation policies and business requirements into structured, machine-executable rules and operational strategies.
• Monitor and analyze model performance metrics (e.g., Accuracy, Recall, Precision, FPR) to identify optimization opportunities.
• Conduct root cause analysis on false positives/negatives and edge cases; propose improvements.
• Support model evaluation, Golden Set construction, and strategy validation to improve model quality and consistency.
• Explore AI-powered automation using LLMs, AI Agents, and workflow automation for analysis and optimization.
• Collaborate with Policy, Product, Engineering, ML, and Operations to drive continuous model improvement.

Skills

Analytical thinking

Cross-functional collaboration

Excel

SQL

Python

Education

Undergraduate/Master's in CS/AI/Data Science/Math

Tools

Excel

SQL

Python

Job description

TikTok is offering an AI Model and Agent Operation Project Intern position (Business Integrity) starting in 2026 in Singapore. You will translate moderation policies into machine-executable rules and collaborate across Policy, Product, Engineering, ML, and Operations to drive model improvements.

You will monitor model metrics, perform root-cause analysis on errors, and support evaluation and strategy validation to enhance model quality and consistency.