AI Agent Operations Specialist （Model Data Operations) - TikTok Platform Safety

TikTok
Singapore
Full timeFunction: Data Science & Machine LearningExperience: 1+ yearsEducation: bachelorsSkills: ["Data analysis","Root cause analysis","Stakeholder management","Problem-solving","Self-motivation"]

Lead operations for an AI agent that supports TikTok Platform Safety, refining prompt strategy and optimizing few-shot samples to improve precise recall for risky content. Perform root-cause analysis on false negatives and false positives, adjust prompts quickly during public opinion and attack surges, and turn short-term strategies into durable capabilities. Maintain a RAG knowledge base, monitor key metrics daily, and deliver daily/weekly operation reports with troubleshooting for abnormal fluctuations.

Loading

Loading job details...

Preparing the role view and application actions.

FursaFursa
TikTok
TikTok
1 month ago

AI Agent Operations Specialist （Model Data Operations) - TikTok Platform Safety

✓ Verified Job

Canonical indexed version, validated from employer's careers page.

Source: Company careers pageValidated by: Fursa AI
Last checked: 8 hours agoStatus: Live

Job Summary

Lead operations for an AI agent that supports TikTok Platform Safety, refining prompt strategy and optimizing few-shot samples to improve precise recall for risky content. Perform root-cause analysis on false negatives and false positives, adjust prompts quickly during public opinion and attack surges, and turn short-term strategies into durable capabilities. Maintain a RAG knowledge base, monitor key metrics daily, and deliver daily/weekly operation reports with troubleshooting for abnormal fluctuations.
Location: Singapore
Employment Type: Full time
Job Function: Data Science & Machine Learning

Key Responsibilities

  • •Write, debug, and maintain prompts for specific risk scenarios, including managing a prompt version library.
  • •Filter and maintain few-shot sample libraries, ensuring coverage of latest underground variants and edge cases.
  • •Analyze false negative/false positive cases to identify causes such as ambiguous prompt instructions, outdated samples, or policy ambiguity.
  • •Adjust prompt strategies quickly in response to public opinion events or new underground attack types, and convert temporary strategies into long-term capabilities.
  • •Maintain the RAG external knowledge base for political/current affairs and person-related risks; monitor/evaluate metrics daily and produce daily/weekly operation reports with timely troubleshooting.

Key Requirements

  • •Bachelor’s degree (or above) with 1+ years of experience in content risk control, strategy operations, search suggestion strategy, or AI product operations.
  • •Deep understanding of content safety and demonstrated success improving model performance through prompt engineering and adjusting context, persona, and examples.
  • •Practical experience with LLM principles and prompt engineering, including writing and debugging prompts for risk scenarios.
  • •Excellent data analysis skills to identify strategic issues through data fluctuations and metric changes.
  • •Self-motivated, results-driven, and able to influence stakeholders in fast-paced, ambiguous environments.
Experience:1+ yearsContent safetyAILLM
Education:Bachelor's
Skills:Data analysisRoot cause analysisStakeholder managementProblem-solvingSelf-motivation
Tech Stack:LLMPrompt engineeringRAGFew-shot examplesData pipelinesKnowledge base

Company Brief

TikTok
Short-form video platform that lets users create, share, and discover entertainment content through algorithmic recommendations. It also offers advertising and creator tools for brands, influencers, and businesses.
Industry: Digital Media
Company Size: Enterprise (1,001+ employees)
Growth: Scaleup
Headquarters: Singapore, Singapore
Founded: 2016
WebsiteLinkedIn