Seung Kwon Cha

Key Skills

Software

Appen

Labelbox

OneForma

Remotasks

Telus

Other

Scale AI

Top Subject Matter

No subject matter listed

Top Data Types

Audio

Image

Text

Top Label Types

Audio Recording

Data Collection

Evaluation Rating

Prompt Response Writing SFT

RLHF

Freelancer Overview

AI Training and Evaluator OneForma/Alignerr/Appen | Apr 2025 – Present - Evaluated and ranked prompt responses from Large Language Models - Assessed quality checks of AI-generated outputs - Conducted reviews of analyst contributions - Revised and composed prompts based on relevance and factual accuracy - Korean transcribing based on audio recording and quality checks AI Training Skills 1. Core Competencies Outlier & Anomaly Detection Expertise in identifying and analyzing outliers in structured/unstructured datasets. Hands-on experience with ML/statistical outlier detection methods Proven ability to clean noisy data and improve model performance through removal. Data Annotation & Labeling High-precision labeling and QA for large-scale datasets (images, text, audio, etc.). Proficiency with annotation tools (Audio collector) Experience establishing labeling guidelines and ensuring consistency for AI training. 2. Technical Skills AI/ML Knowledge: Data preprocessing (scaling, feature engineering), model training/evaluation pipelines. Automation: Scripting for efficiency (e.g., Open CV for image annotation checks). 3. Soft Skills Collaborative work with data scientists/engineers to align labeling with model requirements. Meticulous attention to detail for high-quality training data. Agile workflow management to meet project deadlines.

Labeling Experience

Prompt Response Evaluation Project

Scale AITextEvaluation Rating

Expertise in identifying and analyzing outliers in structured/unstructured datasets. Hands-on experience with ML/statistical outlier detection methods Proven ability to clean noisy data and improve model performance through removal.

Expertise in identifying and analyzing outliers in structured/unstructured datasets. Hands-on experience with ML/statistical outlier detection methods Proven ability to clean noisy data and improve model performance through removal.

2024

Korean Audio Transcription Project

Scale AIAudioAudio Recording

The project is to transcribe the Korean including analyzing the concept of the audio and correcting the transcriptions for having pauses and other details of audio.

The project is to transcribe the Korean including analyzing the concept of the audio and correcting the transcriptions for having pauses and other details of audio.

2025 - 2025

Education

Work History