Samuel A.

Key Skills

Software

Appen

Remotasks

Top Subject Matter

LLM evaluation and RLHF data annotation for commercial AI products

Nlp Domain Expertise

dialogue annotation

Top Data Types

Text

Image

Top Task Types

Entity (NER) Classification

Segmentation

Text Generation

Data Collection

Freelancer Overview

AI Data Trainer & Prompt Evaluator (Outlier AI, Scale AI partner). Brings 11+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Outlier AI, Appen, and Remotasks. Education includes Bachelor of Education, University of Ilorin (2014). AI-training focus includes data types such as Text and Image and labeling workflows including Evaluation, Rating, and Entity (NER) Classification.

Labeling Experience

AI Data Trainer & Prompt Evaluator (Outlier AI, Scale AI partner)

Text

Assessed and ranked English-language AI responses based on quality, helpfulness, factual accuracy, and adherence to instructions. Wrote seed and adversarial prompts to test LLM behavior under edge cases to improve safety and robustness. Performed RLHF-style rating and comparison of model outputs, including rewriting responses to better match human preferences. • Evaluated instruction-following and factuality for quality assurance. • Created prompts to surface failure modes and improve model alignment. • Rated, compared, and rewrote outputs for preference alignment (RLHF). • Contributed to fine-tuning datasets for instruction-following models.

2024 - Present

AI Feedback Contributor (DataAnnotation.tech)

TextText Generation

Provided structured written feedback on AI-generated content, including code, essays, and creative writing, to support generative AI model improvement. Evaluated outputs for coherence, factuality, tone, and alignment with given instructions. Authored comparative preference annotations to support better model response ranking and quality. • Reviewed and wrote feedback for code and writing outputs. • Assessed coherence, factuality, tone, and instruction adherence. • Created comparative preference annotations for ranking. • Supported generative AI training with structured evaluations.

2023 - 2023

Senior Annotator & Task Specialist (Appen)

AppenTextEntity (NER) Classification

Performed structured data annotation for NLP and dialogue datasets, including Named Entity Recognition (NER), sentiment analysis, intent classification, and dialogue annotation. Annotated social media content, audio transcripts, and search query data to support training for recommendation and search-ranking algorithms. Maintained high quality scores to achieve Elite Contributor status and supported label consistency through dispute review and adjudication. • Labeled NER, sentiment, intents, and dialogue content. • Annotated social posts, transcripts, and search queries for retrieval/recommendation. • Reviewed and adjudicated annotation disputes to ensure consistency. • Achieved elite contributor performance over multiple months.

2022 - 2023

Image & Text Annotation Contractor (Remotasks, Scale AI)

RemotasksImageSegmentation

Labeled training images for autonomous vehicle and robotics datasets, including bounding boxes and semantic segmentation masks alongside object classification tags. Transcribed and annotated audio clips for speech recognition model training while following strict style guides. Completed hundreds of micro-tasks with consistently high contributor quality. • Annotated bounding boxes and semantic segmentation masks. • Applied object classification tags for robotics/AV datasets. • Transcribed and annotated audio for ASR training. • Delivered high-quality micro-task performance in top contributor tier.

2022 - 2022

Conversational AI Trainer (Toloka AI, Yandex)

TextData Collection

Generated human-like conversation data and dialogue flows for chatbot and virtual assistant training datasets. Validated machine translations between English and Nigerian English/Pidgin to ensure localization accuracy. Performed quality assurance on completed annotation batches by flagging ambiguous or incorrect labels. • Created dialogue flows for conversational AI training data. • Checked translation quality for localization. • QA-reviewed batches and corrected/flagged labeling issues. • Supported consistency of conversational dataset annotations.

2021 - 2022

Education

Work History