Kana Nagai

Key Skills

Software

Labelbox

Top Subject Matter

No subject matter listed

Top Data Types

Audio

Image

Text

Top Label Types

Evaluation Rating

RLHF

Translation Localization

Freelancer Overview

I am a Japanese–English language professional specializing in AI-generated translation review, localization, and post-editing. I am currently engaged in an AI localization project through a major AI data platform, where I review and post-edit AI-generated English–Japanese translations. My work focuses on ensuring semantic accuracy, natural and appropriate Japanese expression, and overall linguistic quality. I regularly identify mistranslations, omissions, unnatural phrasing, and tone issues, and revise outputs to better align with source meaning and localization expectations. In addition, I have experience with speech-to-text transcription projects on AI data platforms, handling Japanese and English audio with close attention to accuracy, formatting rules, and guideline compliance. I also have over five years of experience as an elementary school teacher. In this role, I consistently evaluated written output and provided clear, criteria-based feedback. This background strongly informs my approach to language quality review, helping me maintain objectivity, consistency, and attention to detail in AI training and localization workflows. With native-level Japanese proficiency and advanced professional English reading and writing skills, I am well suited for English–Japanese localization review, AI translation quality checking and post-editing, and language-focused AI data projects that require careful judgment and reliable output quality.

Labeling Experience

Japanese Audio Data Annotation for AI Speech Recognition

LabelboxAudioEntity Ner ClassificationClassification

Annotated Japanese audio recordings for AI speech recognition, including transcription, speaker identification, and labeling of utterance types. Ensured accuracy and consistency by following detailed guidelines and performing quality checks. Reviewed and validated labeled data to improve overall dataset quality.

Annotated Japanese audio recordings for AI speech recognition, including transcription, speaker identification, and labeling of utterance types. Ensured accuracy and consistency by following detailed guidelines and performing quality checks. Reviewed and validated labeled data to improve overall dataset quality.

2025 - 2025

Education

Work History