Donard Nzioki Mumo

Key Skills

Software

Labelbox

Remotasks

Top Subject Matter

Conversational AI

Large Language Models

Top Data Types

Image

Text

Top Task Types

Bounding Box

RLHF

Freelancer Overview

AI Training, LLM Evaluation & Data Annotation Specialist. Core strengths include Remotasks. AI-training focus includes data types such as Text and labeling workflows including RLHF.

Labeling Experience

High-Precision Audio Labeling

AudioTranscription

Annotated 10+ hours of English singing audio at the phoneme level, ensuring temporal alignment within $1–5$ milliseconds for voice synthesis training. Conducted millisecond-precise annotation of pitch and musical note information, aligning vocal performance with symbolic musical scores. Executed high-precision labelling of vocal techniques, including breathing patterns and glottal stops, using spectrogram interpretation.

Annotated 10+ hours of English singing audio at the phoneme level, ensuring temporal alignment within $1–5$ milliseconds for voice synthesis training. Conducted millisecond-precise annotation of pitch and musical note information, aligning vocal performance with symbolic musical scores. Executed high-precision labelling of vocal techniques, including breathing patterns and glottal stops, using spectrogram interpretation.

2020 - 2024

Chatbot Training & Evaluation Project

RemotasksTextRLHF

As an AI Training, LLM Evaluation & Data Annotation Specialist at Remotasks, I conducted large-scale prompt and response review tasks for LLMs. I evaluated, ranked, and improved conversational AI datasets focusing on accuracy, coherence, and safety. My work involved structured quality analysis and reinforcement learning from human feedback for optimizing AI outputs. • Reviewed and rated AI-generated responses utilizing RLHF methodologies. • Performed intent classification, sentiment analysis, and conversation tagging. • Identified bias, toxicity, and hallucinations within model outputs. • Maintained high annotation accuracy and contributed to dataset integrity.

As an AI Training, LLM Evaluation & Data Annotation Specialist at Remotasks, I conducted large-scale prompt and response review tasks for LLMs. I evaluated, ranked, and improved conversational AI datasets focusing on accuracy, coherence, and safety. My work involved structured quality analysis and reinforcement learning from human feedback for optimizing AI outputs. • Reviewed and rated AI-generated responses utilizing RLHF methodologies. • Performed intent classification, sentiment analysis, and conversation tagging. • Identified bias, toxicity, and hallucinations within model outputs. • Maintained high annotation accuracy and contributed to dataset integrity.

2020 - 2024

Data Labeling & Chatbot Training Specialist

LabelboxImageBounding Box

I annotated and labeled large-scale image datasets using bounding boxes and polygon tools to support computer vision model training. My responsibilities included quality checking, reviewing peer annotations, and collaborating with AI engineers to refine guidelines. I delivered a high volume of precise annotations and maintained strong confidentiality standards. • Labeled 20,000+ images for object detection. • Corrected and validated peer-submitted image annotations. • Used Labelbox, CVAT, and SuperAnnotate extensively. • Ensured compliance with data security protocols.

I annotated and labeled large-scale image datasets using bounding boxes and polygon tools to support computer vision model training. My responsibilities included quality checking, reviewing peer annotations, and collaborating with AI engineers to refine guidelines. I delivered a high volume of precise annotations and maintained strong confidentiality standards. • Labeled 20,000+ images for object detection. • Corrected and validated peer-submitted image annotations. • Used Labelbox, CVAT, and SuperAnnotate extensively. • Ensured compliance with data security protocols.

2020 - 2024

Education

Work History