Emmah M.

Key Skills

Software

CVAT

Labelbox

Top Subject Matter

LLM output evaluation and RLHF dataset quality assurance

Generative image evaluation and prompt adherence QA

Annotation QA and classification consistency review

Top Data Types

Text

Image

3D Sensor

Document

Top Task Types

Classification

Polygon

Cuboid

Bounding Box

Segmentation

Tracking

RLHF

Freelancer Overview

AI Response Evaluator & RLHF Specialist (TURING). Brings 4+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Internal, Proprietary Tooling, and CVAT. Education includes Graduate Diploma in Cartography and Geographic Information Systems, Kenya Institute of Surveying and Mapping (2021) and Basic and Advanced Computer Studies, Compaq Computer College (2018). AI-training focus includes data types such as Text, Image, and Geospatial and labeling workflows including Evaluation, Rating, and Classification.

Labeling Experience

AI Response Evaluator & RLHF Specialist

LabelboxImageRLHF

No description provided.

2026 - Present

AI Response Evaluator & RLHF Specialist (TURING)

Text

Performed side-by-side evaluation of AI-generated responses by assessing quality, clarity, reasoning depth, and helpfulness. Ranked and compared model outputs using Reinforcement Learning from Human Feedback (RLHF) principles to support model fine-tuning improvements. Ensured factual accuracy and safety compliance by cross-checking against reliable sources and applying policy constraints. • Evaluated adherence to complex multi-step instructions, user intent, and formatting requirements. • Identified subtle failure patterns including overgeneralization, contradictions, and incomplete reasoning chains. • Provided structured qualitative feedback highlighting strengths and weaknesses in reasoning and instruction-following. • Recorded edge cases and suggested improvements to internal evaluation rubrics to reduce inter-rater disagreement.

2026 - Present

AI Image Quality & Comparison Specialist (IMERIT)

Image

Compared multiple AI-generated images against specific text prompts to assess short-term quality for generative AI models. Evaluated prompt adherence by verifying that requested objects, styles, and spatial relationships were accurately rendered. Identified and tagged visual anomalies such as anatomical distortions, blurred textures, and lighting inconsistencies. • Used project-specific labeling rubrics to maintain consistency across large-scale image datasets. • Benchmarked fine-grained details including text rendering accuracy, perspective depth, and photorealism metrics. • Assessed performance across diverse artistic styles, aspect ratios, and complex multi-subject prompts. • Produced rapid feedback to support model iteration and image-generation refinement.

2026 - 2026

Data Quality Analyst (CLUTTERBOT)

TextClassification

Reviewed completed annotation tasks to detect missing labels, incorrect classifications, inaccurate boundaries, and guideline violations. Calculated error rates and quality scores to determine acceptance, rejection, or rework requirements. Edited annotations to meet client quality thresholds and delivery standards. • Performed root-cause analysis on recurring annotation errors and documented findings for continuous improvement. • Collaborated with annotators to improve consistency and reduce quality issues. • Delivered coaching and feedback to support higher agreement on labeled outputs. • Used calibration exercises to align labeling decisions across the team.

2025 - 2026

LABEL YOUR DATA/SUPPORT YOUR APP

LabelboxImageBounding BoxPolygon

No description provided.

2025 - 2025

Show more

Education

Work History