Chycik Ayu Winata

Key Skills

Software

Data Annotation Tech

Telus

Top Subject Matter

No subject matter listed

Top Data Types

Document

Image

Text

Top Task Types

Evaluation Rating

Prompt Response Writing SFT

Text Generation

Text Summarization

Translation Localization

Freelancer Overview

An experienced AI training contributor with a strong focus on LLM evaluation and personalized content analysis. I have worked on bilingual (English and Indonesian) projects involving prompt assessment, output review, and cultural localization. My attention to detail and linguistic fluency allow me to evaluate AI-generated content for accuracy, tone, and user alignment. Reviewed and refined AI-generated outputs by assessing relevance, factual accuracy, tone, and user intent. In addition to evaluating conversational AI responses, I have assessed the relevance and quality of personalized ads, ensuring they meet platform standards and user expectations. My background in both technical review and content analysis equips me to contribute to projects that require critical thinking, cultural sensitivity, and language precision.

Labeling Experience

Bilingual LLM Evaluation Rater

Data Annotation TechTextClassificationQuestion Answering

In multiple AI evaluation projects, I assessed chatbot and AI-generated responses across five key dimensions: correctness/truthfulness, verbosity, instruction following, localization, and harmfulness. Tasks included evaluating single-turn and multi-turn interactions, crafting prompts to trigger specific errors or behaviors, and ensuring cultural and linguistic appropriateness in Indonesian while adhering to strict quality guidelines.

In multiple AI evaluation projects, I assessed chatbot and AI-generated responses across five key dimensions: correctness/truthfulness, verbosity, instruction following, localization, and harmfulness. Tasks included evaluating single-turn and multi-turn interactions, crafting prompts to trigger specific errors or behaviors, and ensuring cultural and linguistic appropriateness in Indonesian while adhering to strict quality guidelines.

2024

Talk to a Chatbot and Compare Responses, Focus on FACTUALITY Issues (Multi-Turn & Single-Turn))

Data Annotation TechTextClassificationQuestion Answering

The project focused on evaluating AI chatbot responses in both single-turn and multi-turn settings, with emphasis on detecting and categorizing factuality issues. In addition to assessment, tasks included crafting targeted prompts designed to elicit specific types of factual errors, as defined by the project guidelines. Evaluations covered accuracy, coherence, and alignment with user intent in both English and Indonesian. Strict quality measures were followed, including adherence to scoring rubrics and consistency checks across all evaluation cycles.

The project focused on evaluating AI chatbot responses in both single-turn and multi-turn settings, with emphasis on detecting and categorizing factuality issues. In addition to assessment, tasks included crafting targeted prompts designed to elicit specific types of factual errors, as defined by the project guidelines. Evaluations covered accuracy, coherence, and alignment with user intent in both English and Indonesian. Strict quality measures were followed, including adherence to scoring rubrics and consistency checks across all evaluation cycles.

2025 - 2025

Education

Work History