Mansi Thakur

Key Skills

Software

Appen

Clickworker

HiveMind

Lionbridge

OneForma

Scale AI

Toloka

Telus

Other

Top Subject Matter

No subject matter listed

Top Data Types

Audio

Computer Code Programming

Image

Text

Top Task Types

Audio Recording

Classification

Computer Programming/Coding

Emotion Recognition

Evaluation/Rating

Prompt + Response Writing (SFT)

Question Answering

RLHF

Segmentation

Text Generation

Text Summarization

Transcription

Translation/Localization

Freelancer Overview

I am an experienced AI Trainer and Data Annotator specializing in narrative-driven tasks, dialogue annotation, and scenario evaluation for interactive AI systems. My background includes annotating large datasets, evaluating AI-generated responses for tone, character consistency, and logical flow, as well as conducting guideline-based QA to ensure accuracy and immersion in story-driven environments. I have worked extensively with annotation platforms, web-based QA tools, and data formats such as JSON and CSV, and am skilled in using Excel and Google Sheets for data handling. With a strong blend of creative storytelling and analytical rigor, I excel at identifying inconsistencies, refining content, and supporting the development of realistic, engaging AI behaviors. My passion for gaming, world-building, and narrative design drives my commitment to delivering high-quality training data for next-generation AI experiences.

Labeling Experience

Virtual Agent Audio Evaluation – AI Rater

OtherAudioQuestion AnsweringEmotion Recognition

Evaluated Google Virtual Agent audio interactions, rating responses for correctness, relevance, tone, and context adherence. Performed audio transcription and quality checks to ensure accuracy. Assessed speech prosody, emotion, and user experience, providing feedback to AI trainers for model fine-tuning and improved conversational performance. Maintained high-quality standards for consistent, unbiased evaluation.

Evaluated Google Virtual Agent audio interactions, rating responses for correctness, relevance, tone, and context adherence. Performed audio transcription and quality checks to ensure accuracy. Assessed speech prosody, emotion, and user experience, providing feedback to AI trainers for model fine-tuning and improved conversational performance. Maintained high-quality standards for consistent, unbiased evaluation.

2025

English Translation & Accent Data Annotation Specialist

OneformaAudioEmotion RecognitionTranslation Localization

English audio recordings were annotated for translation and accent evaluation to improve AI models' understanding of diverse English accents and speech recognition. This included transcribing English speech with attention to spelling, punctuation, grammar, and context, as well as labeling accent variations such as British, American, Australian, and regional English accents to enhance AI comprehension of global English speakers.

English audio recordings were annotated for translation and accent evaluation to improve AI models' understanding of diverse English accents and speech recognition. This included transcribing English speech with attention to spelling, punctuation, grammar, and context, as well as labeling accent variations such as British, American, Australian, and regional English accents to enhance AI comprehension of global English speakers.

2023

Hindi Translation & Accent Data Annotation Specialist

AppenAudioEmotion RecognitionTranslation Localization

Hindi audio recordings annotated for translation and accent evaluation, improving AI speech recognition and multilingual understanding. Transcribed Hindi speech to text with accuracy in grammar, punctuation, and contextual meaning. Labeled accent variations, pronunciation differences, and dialects to enhance AI's understanding of diverse Hindi speakers.

Hindi audio recordings annotated for translation and accent evaluation, improving AI speech recognition and multilingual understanding. Transcribed Hindi speech to text with accuracy in grammar, punctuation, and contextual meaning. Labeled accent variations, pronunciation differences, and dialects to enhance AI's understanding of diverse Hindi speakers.

2023

AI Data Labeling & Annotation Specialist – NLP & LLM Project

AppenComputer Code ProgrammingQuestion AnsweringRLHF

* Contributed to the Project, a large-scale AI training initiative focused on SQL and Oracle database query generation using natural language prompts. * Created and annotated diverse prompt-response pairs to teach the AI model how to understand and generate SQL queries of varying difficulty levels, from simple SELECT statements to complex JOIN and aggregation queries. * Performed evaluation and rating of AI responses, ensuring alignment with database logic, syntax accuracy, and natural language interpretation.

* Contributed to the Project, a large-scale AI training initiative focused on SQL and Oracle database query generation using natural language prompts. * Created and annotated diverse prompt-response pairs to teach the AI model how to understand and generate SQL queries of varying difficulty levels, from simple SELECT statements to complex JOIN and aggregation queries. * Performed evaluation and rating of AI responses, ensuring alignment with database logic, syntax accuracy, and natural language interpretation.

2023

Education

Work History