Karanja Muruga

Key Skills

Software

Labelbox

Label Studio

Other

CVAT

Top Subject Matter

AI prompt and response evaluation

LLM and scenario-based AI training content

Video captioning and OCR AI training

Top Data Types

Text

Video

Document

Top Task Types

Prompt Response Writing SFT

RLHF

Classification

Data Collection

Bounding Box

Segmentation

Polygon

Object Detection

Text Generation

Question Answering

Text Summarization

Fine Tuning

Evaluation Rating

Freelancer Overview

Based on my extensive experience in data labeling and AI training, I have developed a robust skill set that encompasses data annotation, quality assurance, and content evaluation for AI models. My role as a Data Annotator at RWS Group Supply Chain has honed my ability to meticulously analyze and label data, ensuring accuracy and adherence to guidelines. Additionally, my experience at Turing involved reviewing user prompts and evaluating AI-generated responses, where I provided detailed feedback to enhance model performance. This work required a keen eye for detail and a deep understanding of AI training methodologies, including supervised fine-tuning (SFT) and reinforcement learning from human feedback (RLHF). I'm pursuing a Master’s degree in Data Science at the University of East London, where I am currently enhancing my expertise in data analysis, visualization, and machine learning. My proficiency in tools such as R, Excel, Tableau, and SQL, combined with my background in user experience design, allows me to create intuitive data visualizations and workflows that support effective decision-making. Notable projects include analyzing health data to identify trends in diabetes and hypertension prevalence, as well as creating impactful data dashboards that deliver actionable insights. My comprehensive training and hands-on experience position me uniquely to contribute to AI training and data labeling initiatives effectively.

Labeling Experience

Data Annotator

OtherImageClassification

At RWS Group Supply Chain as a Data Annotator, I performed labeling and annotation work specifically supporting AI and ML tasks. My duties included classifying and annotating images and text data to aid model training and data quality assurance. I supported AI and machine learning projects through ongoing data labeling workflows. • Regularly annotated and classified text datasets for AI. • Supported multiple AI/ML model development tasks. • Ensured compliance with data privacy and quality standards. • Provided labeled data for supervised learning models.

At RWS Group Supply Chain as a Data Annotator, I performed labeling and annotation work specifically supporting AI and ML tasks. My duties included classifying and annotating images and text data to aid model training and data quality assurance. I supported AI and machine learning projects through ongoing data labeling workflows. • Regularly annotated and classified text datasets for AI. • Supported multiple AI/ML model development tasks. • Ensured compliance with data privacy and quality standards. • Provided labeled data for supervised learning models.

2024 - Present

Computer Use Agent Evaluator – AI Training

OtherText

As a Computer Use Agent Evaluator in AI training, I reviewed user prompts and two candidate AI-generated solutions. I wrote the Gold Standard response, completed detailed side-by-side comparisons, and flagged relevant issues, suggesting improvements for the models. My work focused on evaluating output accuracy and safety while providing actionable feedback to improve AI responses. • Conducted comparative analysis of AI-generated text outputs. • Assessed prompt adherence, output clarity, and adherence to guidelines. • Flagged issues and contributed improvement suggestions for AI models. • Ensured consistency and quality of AI-generated responses.

As a Computer Use Agent Evaluator in AI training, I reviewed user prompts and two candidate AI-generated solutions. I wrote the Gold Standard response, completed detailed side-by-side comparisons, and flagged relevant issues, suggesting improvements for the models. My work focused on evaluating output accuracy and safety while providing actionable feedback to improve AI responses. • Conducted comparative analysis of AI-generated text outputs. • Assessed prompt adherence, output clarity, and adherence to guidelines. • Flagged issues and contributed improvement suggestions for AI models. • Ensured consistency and quality of AI-generated responses.

2026 - 2026

AI Trainer (SFT & RLHF)

LabelboxTextPrompt Response Writing SFT

As an AI Trainer specializing in Supervised Fine Tuning (SFT) and Reinforcement Learning from Human Feedback (RLHF), I created and edited diverse content formats for model development. I annotated data using Labelbox and conducted quality reviews on AI outputs. My tasks included collaborating closely with product teams and providing feedback to improve large language models (LLMs). • Created, edited, and annotated AI training content. • Utilized Labelbox for data labeling assignments. • Reviewed and rated AI-generated outputs for accuracy and quality. • Supported alignment of AI models with operational and user-quality needs.

As an AI Trainer specializing in Supervised Fine Tuning (SFT) and Reinforcement Learning from Human Feedback (RLHF), I created and edited diverse content formats for model development. I annotated data using Labelbox and conducted quality reviews on AI outputs. My tasks included collaborating closely with product teams and providing feedback to improve large language models (LLMs). • Created, edited, and annotated AI training content. • Utilized Labelbox for data labeling assignments. • Reviewed and rated AI-generated outputs for accuracy and quality. • Supported alignment of AI models with operational and user-quality needs.

2025 - 2026

English QA – AI Training (RLHF)

OtherVideoRLHF

In my English QA – AI Training role, I reviewed and annotated video captions for contextual and grammatical accuracy. I analyzed caption inconsistencies, provided detailed feedback, and collaborated to ensure adherence to content guidelines. I also reviewed summarized content used for the RLHF process in an OCR AI model. • Ensured caption accuracy and guideline alignment through annotation. • Identified and addressed errors in video captioning. • Evaluated summarized output for OCR model performance. • Maintained high standards through team collaboration and QA processes.

In my English QA – AI Training role, I reviewed and annotated video captions for contextual and grammatical accuracy. I analyzed caption inconsistencies, provided detailed feedback, and collaborated to ensure adherence to content guidelines. I also reviewed summarized content used for the RLHF process in an OCR AI model. • Ensured caption accuracy and guideline alignment through annotation. • Identified and addressed errors in video captioning. • Evaluated summarized output for OCR model performance. • Maintained high standards through team collaboration and QA processes.

2025 - 2025

AI Fundamentals – Data Labeling & Annotation (Training)

OtherTextData Collection

Through the IBM Skills Build Artificial Intelligence Fundamentals course, I learned and practiced data annotation and labeling for supervised machine learning. My work focused on collecting and labeling text data to improve model outcomes. I gained hands-on experience with annotation workflows in supervised AI contexts. • Practiced supervised data collection and annotation processes. • Focused on text datasets for ML training. • Acquired foundational skills in AI and ML data preparation. • Applied knowledge to improve understanding of AI model training.

Through the IBM Skills Build Artificial Intelligence Fundamentals course, I learned and practiced data annotation and labeling for supervised machine learning. My work focused on collecting and labeling text data to improve model outcomes. I gained hands-on experience with annotation workflows in supervised AI contexts. • Practiced supervised data collection and annotation processes. • Focused on text datasets for ML training. • Acquired foundational skills in AI and ML data preparation. • Applied knowledge to improve understanding of AI model training.

2024 - 2024

Education

Work History