Open to opportunities

khotso thelingoane

@khotsothelingoane

AI data annotator and Sesotho sa Leboa linguist, fine-tuning datasets and evaluating model outputs for accuracy and alignment.

South Africa

Message

What I'm looking for

I’m looking for remote, human-in-the-loop AI training work where I can annotate and validate Sesotho sa Leboa data, score outputs by rubrics, and improve generative models with rigorous linguistic quality checks.

I’m an AI data annotator and Sesotho sa Leboa Language Specialist with hands-on experience in AI data annotation, model fine-tuning support, and linguistic validation. I bring native-level fluency in Sesotho sa Leboa and a practical focus on AI alignment through human-in-the-loop workflows.

I’ve curated and annotated AI/ML training datasets for linguistic accuracy, then evaluated AI-generated text using rubric-based scoring (factuality, fluency, and cultural relevance). I train and refine datasets across dialectal variations, slang, and formal vs. informal registers—so models learn the difference between nuance and literal translation.

Alongside annotation, I strengthen content quality assurance through fact-checking, brevity and coherence checks, and bias detection. In my AI Text Evaluation project, I manually scored 1000+ responses and identified common failure modes like grammatical errors and literal translations.

I also contribute to bilingual language work and large-scale corpora: I’ve produced and edited Sesotho sa Leboa/English content with structured guidelines for consistency, and I’ve supported localization/translation teams for AI training datasets. My professional ethos is simple: combine linguistic depth with rigorous evaluation so generative AI outputs are trustworthy, natural, and culturally grounded.

Experience

Work history, roles, and key accomplishments

AI Data Annotator & Linguist

Outlier AI

Jan 2025 - Present (1 year 4 months)

Annotated and curated Sesotho sa Leboa datasets for AI/ML training, ensuring linguistic accuracy across dialects and formal vs. informal registers. Evaluated AI-generated text using rubrics for factuality, fluency, and cultural relevance, providing iterative feedback to improve model performance.

AI Data Annotation LLM Evaluation Linguistic Validation

Content Editor & Writer

Appen International

Jan 2025 - Present (1 year 4 months)

Produced and edited bilingual Sesotho sa Leboa and English content for large-scale corpora, maintaining consistency with structured guidelines. Collaborated with translation/localization teams to adapt content for AI training use.

Localization Corpus Consistency Linguistic Guideline Adherence Fact Checking

Bilingual Corpus Annotation

Freelance

Jan 2024 - Present (2 years 4 months)

Tagged linguistic features (including sentiment and named entities) in Sesotho sa Leboa datasets to support benchmarking for African language NLP. Contributed labeled examples aligned to annotation standards for model evaluation and training.

Linguistic Feature Labeling Benchmarking Inter Annotator Consistency Annotations NLP Data Preparation African Language NLP

AI Text Evaluation (Sesotho)

Freelance

Jan 2024 - Present (2 years 4 months)

Manually scored 1,000+ AI-generated Sesotho sa Leboa responses for fluency, bias, and factual correctness. Identified recurring failure modes such as literal translation artifacts and grammatical issues to inform annotation guidelines.

Fluency Rubric Based QA

Freelance Researcher/Journalist

Freelance

Created structured, fact-checked written materials that could be repurposed for AI training datasets. Produced research and journalism-style content with a focus on accuracy and clarity for generative AI workflows.

Journalism Fact Checking Content Quality Assurance Technical Documentation

Freelance Translator (Multilingual)

Freelance

Translated multilingual datasets with emphasis on semantic accuracy and faithful meaning transfer. Supported downstream AI training by ensuring translations were suitable for model consumption and evaluation.

Linguistic QA Bilingual Alignment Tone Preservation Terminology Management Proofreading