Moses Njehia

Key Skills

Software

CVAT

Dataloop

Labelbox

Label Studio

Scale AI

Other

Top Subject Matter

No subject matter listed

Top Data Types

Text

Image

Audio

Top Label Types

Bounding Box

Classification

Data Collection

Emotion Recognition

Entity Ner Classification

Evaluation Rating

Fine Tuning

Polygon

Red Teaming

Segmentation

Text Generation

Transcription

Freelancer Overview

I am an experienced AI trainer and data annotator with over five years of hands-on involvement in large-scale AI training and evaluation projects. My background includes evaluating conversational and long-form AI outputs for clarity, coherence, contextual alignment, and tone consistency, with a strong focus on dialogue and natural language processing tasks. I have annotated diverse data types—including text, image, and audio—to support the development of robust AI models. I excel at applying complex evaluation rubrics, interpreting detailed guidelines, and maintaining high accuracy in remote, independent settings. My strengths lie in my attention to detail, analytical thinking, and ability to ensure consistency and quality across annotation and evaluation tasks.

Labeling Experience

Conversational AI Evaluation & Multi-Modal Data Annotation for LLM Training

OtherTextText GenerationFine Tuning

This project involved large-scale conversational AI evaluation and multi-modal data annotation to support the training and fine-tuning of large language models (LLMs). I performed structured tasks such as response evaluation and rating using detailed rubrics, pairwise comparison of AI outputs, RLHF-based feedback, named entity recognition (NER), text classification, sentiment labeling, and contextual accuracy assessments. I also annotated image and audio datasets for classification and validation tasks according to defined taxonomies. The project covered thousands of data samples across multiple batches, contributing to iterative model improvement cycles. I adhered to strict quality standards, including consistent guideline application, participation in calibration sessions to improve inter-annotator agreement, internal quality reviews, and maintaining 95%+ accuracy benchmarks while meeting productivity and turnaround targets in a secure remote environment.

This project involved large-scale conversational AI evaluation and multi-modal data annotation to support the training and fine-tuning of large language models (LLMs). I performed structured tasks such as response evaluation and rating using detailed rubrics, pairwise comparison of AI outputs, RLHF-based feedback, named entity recognition (NER), text classification, sentiment labeling, and contextual accuracy assessments. I also annotated image and audio datasets for classification and validation tasks according to defined taxonomies. The project covered thousands of data samples across multiple batches, contributing to iterative model improvement cycles. I adhered to strict quality standards, including consistent guideline application, participation in calibration sessions to improve inter-annotator agreement, internal quality reviews, and maintaining 95%+ accuracy benchmarks while meeting productivity and turnaround targets in a secure remote environment.

2020 - 2025

Audio Transcription & Speech Data Annotation for ASR Training

Scale AIAudioEmotion RecognitionTranscription

Contributed to speech data annotation projects supporting Automatic Speech Recognition (ASR) and voice-enabled AI systems. Performed accurate audio transcription, speaker labeling, timestamp alignment, audio classification, and emotion recognition tasks following strict linguistic and formatting guidelines. Evaluated speech clarity, background noise levels, pronunciation accuracy, and contextual meaning to ensure high-quality dataset preparation. Reviewed and corrected transcripts for spelling, grammar, and consistency while adhering to detailed annotation taxonomies. Worked on large audio batches under productivity targets while maintaining 95%+ accuracy standards. Participated in quality review and calibration processes to improve inter-annotator agreement and ensure reliable training data for conversational and voice AI models.

Contributed to speech data annotation projects supporting Automatic Speech Recognition (ASR) and voice-enabled AI systems. Performed accurate audio transcription, speaker labeling, timestamp alignment, audio classification, and emotion recognition tasks following strict linguistic and formatting guidelines. Evaluated speech clarity, background noise levels, pronunciation accuracy, and contextual meaning to ensure high-quality dataset preparation. Reviewed and corrected transcripts for spelling, grammar, and consistency while adhering to detailed annotation taxonomies. Worked on large audio batches under productivity targets while maintaining 95%+ accuracy standards. Participated in quality review and calibration processes to improve inter-annotator agreement and ensure reliable training data for conversational and voice AI models.

2019 - 2023

Multi-Modal Data Annotation for Computer Vision & NLP Models

Scale AIImageBounding BoxPolygon

Worked on large-scale AI training projects involving image annotation to support machine learning model development. Performed structured labeling tasks including bounding box annotation, object detection, image classification, segmentation, named entity recognition (NER). Reviewed datasets for accuracy, clarity, and compliance with detailed multi-step guidelines while handling high-volume batches of data. Contributed to quality review assignments and peer validation processes to improve annotation consistency and inter-annotator agreement. Maintained strong accuracy standards (95%+) while meeting productivity targets in a deadline-driven remote environment, ensuring reliable and high-quality outputs for computer vision and NLP systems.

Worked on large-scale AI training projects involving image annotation to support machine learning model development. Performed structured labeling tasks including bounding box annotation, object detection, image classification, segmentation, named entity recognition (NER). Reviewed datasets for accuracy, clarity, and compliance with detailed multi-step guidelines while handling high-volume batches of data. Contributed to quality review assignments and peer validation processes to improve annotation consistency and inter-annotator agreement. Maintained strong accuracy standards (95%+) while meeting productivity targets in a deadline-driven remote environment, ensuring reliable and high-quality outputs for computer vision and NLP systems.

2019 - 2023

Education

Work History