Benson Tenai

Key Skills

Software

Anno-Mage

CloudFactory

CVAT

Data Annotation Tech

Deep Systems

Mindrift

Remotasks

Scale AI

Snorkel AI

VoTT

Doccano

Top Subject Matter

No subject matter listed

Top Data Types

Computer Code Programming

Document

Text

Top Task Types

Action Recognition

Classification

Diagnosis

Fine Tuning

Prompt Response Writing SFT

Freelancer Overview

I am an experienced language specialist and AI training contributor with a strong background in data labeling, prompt engineering, and multilingual LLM evaluation. With over four years of freelance experience in academic writing, content creation, and linguistic analysis, I bring deep attention to detail, consistency, and contextual understanding especially in low-resource and African languages such as Kiswahili and Kalenjin. My strengths lie in text classification, sentiment analysis, entity recognition, and content evaluation for LLMs.

Labeling Experience

Multilingual LLM Evaluation and Text Classification for African Language Models

DoccanoTextEntity Ner ClassificationText Generation

This project involved labeling and evaluating a multilingual dataset designed to improve large language model (LLM) performance in low-resource African languages, specifically Kiswahili and Kalenjin. Tasks included sentiment tagging, named entity recognition, prompt-response evaluations, translation quality checks, and rating AI-generated text for accuracy, fluency, and cultural appropriateness. I was also involved in refining prompt engineering strategies to train and test generative models in diverse linguistic contexts. The project included over 15,000 labeled text samples and adhered to strict quality control measures, including double-pass reviews and inter-annotator agreement checks.

This project involved labeling and evaluating a multilingual dataset designed to improve large language model (LLM) performance in low-resource African languages, specifically Kiswahili and Kalenjin. Tasks included sentiment tagging, named entity recognition, prompt-response evaluations, translation quality checks, and rating AI-generated text for accuracy, fluency, and cultural appropriateness. I was also involved in refining prompt engineering strategies to train and test generative models in diverse linguistic contexts. The project included over 15,000 labeled text samples and adhered to strict quality control measures, including double-pass reviews and inter-annotator agreement checks.

2024 - 2025

Education

Work History