Esther Githiomi

Key Skills

Software

Label Studio

Snorkel AI

Top Subject Matter

No subject matter listed

Top Data Types

Audio

Text

Top Label Types

Audio Recording

Bounding Box

Classification

Entity Ner Classification

Evaluation Rating

Prompt Response Writing SFT

Text Summarization

Freelancer Overview

I am a detail-oriented data annotation and linguistic specialist with over 5 years of experience supporting AI and NLP projects. My expertise spans high-accuracy data labeling, text classification, named entity recognition, sentiment analysis, and RLHF (Reinforcement Learning from Human Feedback) for both English and Swahili datasets. I have a proven record of improving model accuracy by performing deep linguistic audits, identifying hallucinations in AI-generated content, and refining annotation guidelines in close collaboration with engineering teams. My background in legal analysis allows me to provide nuanced insights, especially in technical and regulatory domains. I am highly proficient with tools such as Labelbox, Prodigy, Scale AI, and Appen, and have consistently delivered top-quality results, including acting as a gold standard reviewer on large-scale projects. My strong analytical reasoning, ethical bias detection, and project management skills ensure that I deliver reliable, culturally relevant, and bias-mitigated training data for advanced AI systems.

Labeling Experience

Data Annotation and Quality Review Practice Project (Computer Vision and NLP)

Label StudioTextBounding BoxEntity Ner Classification

Completed structured self-directed data labeling and annotation practice involving image and text datasets to build proficiency in AI training workflows. Tasks included image classification, bounding box annotation for object detection, and named entity recognition (NER) on document datasets. Applied detailed quality control standards such as consistency checks, guideline adherence, and error review to ensure accurate annotations. Worked with sample datasets to understand labeling taxonomies, ambiguity handling, and edge-case identification. Developed strong attention to detail, instruction compliance, and fast turnaround while maintaining labeling accuracy, skills transferable from legal document review and research work.

Completed structured self-directed data labeling and annotation practice involving image and text datasets to build proficiency in AI training workflows. Tasks included image classification, bounding box annotation for object detection, and named entity recognition (NER) on document datasets. Applied detailed quality control standards such as consistency checks, guideline adherence, and error review to ensure accurate annotations. Worked with sample datasets to understand labeling taxonomies, ambiguity handling, and edge-case identification. Developed strong attention to detail, instruction compliance, and fast turnaround while maintaining labeling accuracy, skills transferable from legal document review and research work.

2025

Education

Work History