For employers

Hire this AI Trainer

Sign in or create an account to invite AI Trainers to your job.

LoginCreate account
Invite to Job
Riga Al Amin

Riga Al Amin

AI/LLM & Content QA Expert - Indonesian Trust & Safety Specialist

Indonesia flagJakarta, Indonesia
$8.00/hrExpertOtherInternal Proprietary Tooling

Key Skills

Software

Other
Internal/Proprietary Tooling

Top Subject Matter

No subject matter listed

Top Data Types

ImageImage
TextText
VideoVideo

Top Task Types

Action Recognition
Classification
Emotion Recognition
Evaluation Rating
Prompt Response Writing SFT

Freelancer Overview

I am a native Indonesian language specialist with over seven years of experience in trust and safety, policy enforcement, and content quality across APAC. Since 2016, I have worked in market specialisation for the trust and safety industry, covering customer operations, content moderation, quality analyst, and policy enforcement (Policy Gen AI) while supporting AI and LLM evaluation projects. My expertise includes prompt testing, annotation, data labelling, and error analysis, along with practical use of NLP tools such as TTS, ASR, OCR, and NLU to assess and improve AI outputs across text, audio, and video. I combine linguistic knowledge with systematic testing to ensure models deliver responses that are factual, fluent, and culturally accurate. At TikTok, I improved AI model performance by curating Indonesian slang, dialects, and new terms, reducing undetected violations by 15 percent and maintaining 99% accuracy in reviews. I managed more than 1,000 high-risk and government-sensitive escalations tied to elections, scams, misinformation, and exploitation, achieving 95% SLA compliance under strict deadlines. Beyond escalations, I designed calibration sessions, training playbooks, and SOPs that improved inter-rater alignment, raised policy comprehension by over 90%, and cut error rates by 40%. These experiences highlight my ability to bridge human expertise and AI development, providing reliable evaluation data and ensuring models accurate, relevance, and trustworthy.

ExpertMalayArabicEnglishIndonesian

Labeling Experience

Policy Generative AI (Policy Enforcement)

Internal Proprietary ToolingVideoAction RecognitionDiagnosis
As Policy Enforcement Specialist for the APAC market, I supported AI moderation and LLM initiatives by testing prompts, identifying vulnerabilities, and refining outputs to reduce false negatives by 1.5%. I reviewed complex video content using NLP tools such as TTS, ASR, OCR, and NLU, improving harmful content detection accuracy by 20%. My responsibilities included large-scale data labelling and annotation of 10,000+ cases, tagging false positives and false negatives, which lowered global error rates below 10%. I also expanded AI linguistic coverage by curating Indonesian slang, dialects, and emerging vocabulary, reducing undetected violations by 15% in the local market.

As Policy Enforcement Specialist for the APAC market, I supported AI moderation and LLM initiatives by testing prompts, identifying vulnerabilities, and refining outputs to reduce false negatives by 1.5%. I reviewed complex video content using NLP tools such as TTS, ASR, OCR, and NLU, improving harmful content detection accuracy by 20%. My responsibilities included large-scale data labelling and annotation of 10,000+ cases, tagging false positives and false negatives, which lowered global error rates below 10%. I also expanded AI linguistic coverage by curating Indonesian slang, dialects, and emerging vocabulary, reducing undetected violations by 15% in the local market.

2023 - 2025

AI/LLM Safety Moderation Audit Project

OtherVideoText GenerationObject Detection
At TikTok, I specialised in video-based AI moderation through large-scale data labelling projects for trust and safety evaluation. My scope included checking ASR, OCR, and quality labelling according to the violation depicted. The project involved reviewing final result that has been moderated by our safety agent, my task is to ensure the audit accuracy meet minimum quality target 95% benchmark for AI LLM Moderation support project.

At TikTok, I specialised in video-based AI moderation through large-scale data labelling projects for trust and safety evaluation. My scope included checking ASR, OCR, and quality labelling according to the violation depicted. The project involved reviewing final result that has been moderated by our safety agent, my task is to ensure the audit accuracy meet minimum quality target 95% benchmark for AI LLM Moderation support project.

2023 - 2025

AI Comment Report Labelling

Internal Proprietary ToolingTextAction RecognitionTracking
Identification of comment report labelling and annotation, handling cases on daily basis while working as Quality Analyst & Operations specialist. Tasks including audit comment reports, correct tag labelling, and began in 2023 launched project for prompt testing related to comment labelling on TikTok Platform. During position as SME QA & PES, enforcing localised solutions for ID market through alignment and monitoring quality performance. Quality Benchmarks minimum 95%.

Identification of comment report labelling and annotation, handling cases on daily basis while working as Quality Analyst & Operations specialist. Tasks including audit comment reports, correct tag labelling, and began in 2023 launched project for prompt testing related to comment labelling on TikTok Platform. During position as SME QA & PES, enforcing localised solutions for ID market through alignment and monitoring quality performance. Quality Benchmarks minimum 95%.

2021 - 2025

Subject Matter Experts of Quality Labelling Analyst

Internal Proprietary ToolingVideoAction RecognitionTracking
As a Subject Matter Expert of Quality Analyst (SME QA) at TikTok, I ensured high-quality data labelling and QA standards for the Indonesian market, consistently achieving 98.82% accuracy by adapting global policies into localised guidelines. I acted as the final arbitrator for 20+ daily appeals, rewriting and clarifying ambiguous cases to cut valid errors by 30% each quarter. I also worked with global policy teams to identify gaps and misalignments, reducing recurring issues by 25%. In addition, I developed performance improvement plans that raised underperforming analysts’ accuracy to 98% within three weeks and delivered training programs that enabled new hires to achieve 95%+ accuracy during go-live.

As a Subject Matter Expert of Quality Analyst (SME QA) at TikTok, I ensured high-quality data labelling and QA standards for the Indonesian market, consistently achieving 98.82% accuracy by adapting global policies into localised guidelines. I acted as the final arbitrator for 20+ daily appeals, rewriting and clarifying ambiguous cases to cut valid errors by 30% each quarter. I also worked with global policy teams to identify gaps and misalignments, reducing recurring issues by 25%. In addition, I developed performance improvement plans that raised underperforming analysts’ accuracy to 98% within three weeks and delivered training programs that enabled new hires to achieve 95%+ accuracy during go-live.

2022 - 2023

E-commerce Quality and Annotation

Internal Proprietary ToolingVideoAction RecognitionTracking
At TikTok, I worked on AI e-commerce safety projects for TikTok Shop, focusing on labelling violations such as undisclosed marketing, visual depictions of restricted items, and unsafe products sold by sellers. The scope included daily reviews of large case volumes using ASR, OCR, and NLU to detect violations across multimedia content. I supported monetisation safety by providing accurate annotations, consistently exceeding the minimum 95% quality benchmark while ensuring reliable data for AI model training.

At TikTok, I worked on AI e-commerce safety projects for TikTok Shop, focusing on labelling violations such as undisclosed marketing, visual depictions of restricted items, and unsafe products sold by sellers. The scope included daily reviews of large case volumes using ASR, OCR, and NLU to detect violations across multimedia content. I supported monetisation safety by providing accurate annotations, consistently exceeding the minimum 95% quality benchmark while ensuring reliable data for AI model training.

2021 - 2023

Education

U

University Of Malaya

Master Of Arts, International Relations

Master Of Arts
2015 - 2016
U

University Of Malaya

Bachelor Of Arts, International Relations

Bachelor Of Arts
2011 - 2014

Work History

T

TDCX Thailand

Community Operations Specialist - Indonesian Market

Bangkok
2018 - 2021
S

Sutherland Global Services

CSTS Analyst (Customer Support) - Indonesian Market

Kuala Lumpur
2016 - 2017