Thadius N.

Key Skills

Software

Scale AI

Remotasks

Surge AI

Top Subject Matter

Large language model (LLM) training data

safety/toxicity and preference ranking

Computer vision training data (object localization, segmentation, image captioning)

Top Data Types

Text

Image

Document

Top Task Types

Classification

Bounding Box

Freelancer Overview

Freelance Data Labeling Specialist & AI Trainer. Brings 2+ years of professional experience across legal operations, contract review, compliance, and structured analysis. Core strengths include Scale AI, Remotasks, and Surge AI. Education includes Bachelor of Science, University of Nairobi (2022). AI-training focus includes data types such as Text and Image and labeling workflows including Classification, Bounding Box, and Evaluation.

Labeling Experience

Freelance Data Labeling Specialist & AI Trainer

Scale AIImageBounding Box

Performed image annotation work for computer vision training datasets as part of LLM/AI training data preparation. Drew bounding boxes, created semantic segmentation labels, and produced scene description captions used for model learning. Ensured quality by detecting guideline-ambiguous edge cases and supporting QA updates with clear written rationales. • Produced bounding box annotations for relevant objects • Completed semantic segmentation labeling for visual regions • Authored scene description captions for image-to-text training • Identified and escalated ambiguous labeling edge cases for standard updates

2023 - Present

Freelance Data Labeling Specialist & AI Trainer

Scale AITextClassification

Served as a freelance data labeling specialist and AI trainer performing LLM training data annotation across multiple task types. Completed text classification, entity tagging, coreference resolution, and toxic content detection following detailed guidelines and style rubrics. Contributed to RLHF preference ranking by selecting/scoring response pairs for safety, helpfulness, and factual accuracy. • Labeled multi-class text classification and entity tagging datasets • Performed coreference resolution and toxic content detection annotations • Conducted RLHF preference ranking on AI response pairs • Escalated 200+ ambiguous edge cases with written rationales to QA leads

2023 - Present

Data Quality Reviewer & Content Annotator

Text

Worked as a data quality reviewer and content annotator focused on improving label consistency in existing text datasets. Reviewed low inter-rater agreement records, re-labeled them, and produced correction reports used to retrain internal annotators. Annotated social media content for sentiment polarity, topic category, and policy-violation flags using a detailed moderation taxonomy in English and Swahili. • Performed re-labeling to raise inter-rater agreement on flagged records • Labeled sentiment polarity and topic category for social media text • Applied policy-violation flags using an English/Swahili moderation taxonomy • Maintained structured annotation logs and contributed to label taxonomy expansion

2022 - 2022

Education

Work History