Yatharth Khanna

Key Skills

Software

Labelbox

Mindrift

Top Subject Matter

No subject matter listed

Top Data Types

Computer Code Programming

Image

Text

Top Label Types

Question Answering

Text Summarization

RLHF

Evaluation Rating

Computer Programming Coding

Freelancer Overview

I am a passionate AI-ML enthusiast with hands-on experience in data labeling, annotation, and building AI training datasets across domains like computer vision and NLP. My background includes implementing ML models for sentiment analysis, image classification (CIFAR-10), and developing an AI-driven cybersecurity platform, where I handled data preprocessing, annotation, and validation to ensure high model accuracy. I am proficient in Python (TensorFlow, Keras, Pandas, Scikit-learn, OpenCV, NLTK, spaCy), SQL, and have worked extensively with data pipelines and APIs. My internships and projects have strengthened my skills in preparing, cleaning, and structuring data for machine learning, and I am committed to delivering high-quality, reliable datasets to power intelligent systems.

Labeling Experience

LLM Data Annotation & AI Response Evaluation

MindriftTextQuestion AnsweringText Summarization

Worked as a contributor on AI training data projects involving evaluation and annotation of large language model outputs. Tasks included reviewing conversation transcripts, rating model responses based on relevance, accuracy, and safety, and generating concise summaries of dialogues. Performed text classification and response quality evaluation aligned with Reinforcement Learning from Human Feedback (RLHF) guidelines to help improve model performance. Ensured adherence to annotation instructions and maintained high quality standards while completing micro-tasks across different AI training datasets.

Worked as a contributor on AI training data projects involving evaluation and annotation of large language model outputs. Tasks included reviewing conversation transcripts, rating model responses based on relevance, accuracy, and safety, and generating concise summaries of dialogues. Performed text classification and response quality evaluation aligned with Reinforcement Learning from Human Feedback (RLHF) guidelines to help improve model performance. Ensured adherence to annotation instructions and maintained high quality standards while completing micro-tasks across different AI training datasets.

2025

Education

Work History