Miriam O.

Key Skills

Software

Label Studio

Top Subject Matter

Advanced STEM (Physics, Mathematics, Chemistry)

Business analytics and professional reporting

Agentic computer-use workflows and tool execution

Top Data Types

Text

Image

Document

Top Task Types

RLHF

Function Calling

Freelancer Overview

AI Trainer / Subject Matter Expert (SME) at Turing (Remote). Brings 8+ years of professional experience across legal operations, contract review, compliance, and structured analysis. Core strengths include Populous Studio and Label Studio. Education includes Master of Business Administration, Nexford University (2025) and Bachelor of Laws, University of Abuja (2017). AI-training focus includes data types such as Text, Computer Code, and Programming and labeling workflows including RLHF, Function Calling, and Evaluation.

Labeling Experience

AI Trainer / Subject Matter Expert (SME) — Project: MM-vision-Oracle-ComplexData

Label StudioImage

Trained and evaluated multimodal vision models by using Label Studio and Populous Studio to interpret dense visual elements. Labeled and verified model understanding of scientific diagrams, charts, structural schemas, and complex data tables to ensure correct logical analysis. Improved multimodal model performance by correcting interpretation errors tied to visual complexity. • Prepared and validated multimodal training inputs in Label Studio/Populous Studio. • Assessed diagram/chart/table comprehension and logical extraction quality. • Ensured outputs matched expected interpretations and structured schema understanding. • Iteratively refined labeling to address missed or incorrect visual reasoning.

2025 - Present

AI Trainer / Subject Matter Expert (SME) — Project: GOOGLE_GDM_Computer_use Evals

Label StudioFunction Calling

Labeled and evaluated datasets for agentic computer-use workflows, assessing tool execution, interface navigation, and autonomous task planning within computer environments. Identified nuanced failure modes, tool-execution gaps, and boundary issues to improve agent reliability and overall workflow performance. Documented results through structured dataset labeling in agent evaluation projects. • Evaluated agent ability to execute complex operations in UI-based environments. • Tracked and labeled tool-use execution behavior and planning outcomes. • Diagnosed failure modes to guide reliability improvements. • Assessed performance while agents operated across Google Workspace and desktop apps.

2025 - Present

AI Trainer / Subject Matter Expert (SME) at Turing (Remote)

TextRLHF

Trained and evaluated LLM responses using RLHF-style quality improvement workflows in Populous Studio and Label Studio environments, focusing on logical correctness, factuality, and rubric compliance. Performed iterative review, ranking, and correction of model outputs for enterprise business contexts including reports, executive summaries, and strategy briefs. Worked as a Subject Matter Expert to ensure responses adhered to complex academic and professional standards. • Evaluated and optimized AI-generated responses against detailed quality rubrics. • Conducted response ranking and improvement cycles to enhance safety and factuality. • Reviewed multi-step reasoning for consistency, completeness, and guideline adherence. • Produced high-quality reference-style business prose and structured summaries for training use.

2025 - Present

Education

Work History