Mutuma M.

Key Skills

Software

Remotasks

Mercor

Top Subject Matter

AI assistant response evaluation and data annotation for NLP/LLM training

AI training data labeling and categorization (text)

LLM/AI evaluation

Top Data Types

Text

Video

Top Task Types

Classification

Polygon

Segmentation

Freelancer Overview

Data Annotation Specialist (AI evaluation and labeling), DataAnnotation (Remote). Core strengths include DataAnnotation, Remotasks, and Mercor. Education includes Doctor of Philosophy, University of Nairobi (2000) and Bachelor of Science, Kenyatta University (1996). AI-training focus includes data types such as Text and labeling workflows including Evaluation, Rating, and Classification.

Labeling Experience

AI Trainer / Data Annotator (AI response evaluation), Outlier AI, Mercor AI / Handshake Projects (Remote)

MercorText

Reviewed and evaluated AI-generated responses for accuracy, clarity, relevance, and instruction-following according to project guidelines. Provided structured feedback to improve model performance across general, technical, and reasoning-based tasks. Analyzed prompts and responses to identify factual errors, weak reasoning, formatting issues, and incomplete answers while completing remote projects independently. • Assessed response quality across accuracy, clarity, relevance, and instruction adherence • Delivered structured model feedback and reasoning checks • Performed prompt/response analysis for factuality and formatting issues • Completed high-quality remote evaluations aligned with platform deadlines

2023 - Present

Data Annotation Specialist (AI evaluation and labeling), DataAnnotation (Remote)

Text

Evaluated AI assistant responses using detailed rubrics and scoring guidelines to assess helpfulness, factuality, completeness, and tone. Compared model outputs to determine which response was stronger and maintained scoring consistency across assignments. Performed multi-step reasoning, writing, editing, and classification tasks as part of online AI evaluation work. • Applied rubric-based evaluation criteria to judge response quality • Compared multiple model outputs to select the preferred answer • Completed multi-step reasoning and classification labeling • Ensured accuracy and consistency across diverse online tasks

2021 - 2023

Remote Task Contributor (data labeling and review), Remotasks (Remote)

RemotasksTextClassification

Completed online data labeling, categorization, and review tasks for AI training projects using structured instructions. Annotated information accurately for remote tasks and performed quality checks to correct inconsistencies when found. Worked in a fast-paced remote environment focused on producing labeled outputs suitable for model training. • Labeled and categorized data according to written instructions • Reviewed labeling outputs and corrected inconsistencies • Supported AI training data preparation through remote tasks • Maintained quality while working on time-bound assignments

2019 - 2022

Education

Work History