David Muigai

Key Skills

Software

AWS SageMaker

Data Annotation Tech

Labelbox

Remotasks

Scale AI

Top Subject Matter

No subject matter listed

Top Data Types

Computer Code Programming

Text

Video

Top Task Types

Action Recognition

Audio Recording

Bounding Box

RLHF

Freelancer Overview

I have 5 years of experience in data labeling and AI training, with a focus on [industry/subject area, e.g., natural language processing, computer vision, etc.]. My work has involved tasks such as [annotation types – e.g., text classification, entity recognition, image segmentation, sentiment tagging, etc.], ensuring high accuracy and consistency across large datasets. I am proficient with labeling tools such as [tools you've used – e.g., Labelbox, Prodigy, Scale AI, etc.], and I understand the importance of clean, well-structured data for training robust AI models. What sets me apart is my [key strength – e.g., linguistic background, technical knowledge, attention to detail, fluency in other languages, etc.], which allows me to contribute effectively to projects involving nuanced decision-making and subject matter expertise. Whether working independently or as part of a team, I bring a high level of precision, accountability, and curiosity to every task

Labeling Experience

LLM Text Data Evaluation and Annotation for NLP Training

LabelboxTextEntity Ner ClassificationText Generation

Worked on a large-scale LLM training project involving the annotation and evaluation of English text data to improve model quality and alignment. Responsibilities included classifying prompts, evaluating model-generated responses for factuality, coherence, and helpfulness, and providing written feedback for reinforcement learning with human feedback (RLHF). Also participated in prompt-response generation for supervised fine-tuning (SFT). The project required consistently high levels of attention to detail, understanding of nuanced language, and adherence to strict quality assurance guidelines. Delivered over 15,000+ labeled examples with a 98% reviewer acceptance rate.

Worked on a large-scale LLM training project involving the annotation and evaluation of English text data to improve model quality and alignment. Responsibilities included classifying prompts, evaluating model-generated responses for factuality, coherence, and helpfulness, and providing written feedback for reinforcement learning with human feedback (RLHF). Also participated in prompt-response generation for supervised fine-tuning (SFT). The project required consistently high levels of attention to detail, understanding of nuanced language, and adherence to strict quality assurance guidelines. Delivered over 15,000+ labeled examples with a 98% reviewer acceptance rate.

2023

Education

Work History