Lawrencia G.

Key Skills

Software

Don't disclose

Top Subject Matter

Multilingual AI training (Hausa/English) and support-response evaluation

LLM evaluation

RLHF-style preference quality

Top Data Types

Text

Top Task Types

RLHF

Freelancer Overview

Virtual Tech Assistant (Freelance/Remote). Brings 7+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Don't disclose, Zendesk, and Freshdesk. Education includes Bachelor of Science, Kaduna State University. AI-training focus includes data types such as Text and labeling workflows including Evaluation, Rating, and RLHF.

Labeling Experience

Ongoing evaluation and quality-assurance work in a remote support environment

TextRLHF

Conducted human-quality evaluation of customer-support interactions that functionally parallels RLHF preference data collection and review. Focused on identifying issues in response quality, surfacing errors, and ensuring responses addressed user needs accurately. Used consistent evaluation criteria to support scalable feedback and annotation quality for AI training. • Checked responses for helpfulness, completeness, and correctness • Identified errors and gaps relevant to hallucination-like failures • Maintained high satisfaction metrics via quality assurance • Enabled repeatable evaluation criteria through documentation

2020 - Present

Virtual Tech Assistant (Freelance/Remote)

Don't discloseText

Evaluated and triaged support-ticket responses by urgency, quality, and resolution accuracy to ensure response correctness and usefulness. Applied structured response quality assessment practices that align with LLM evaluation and RLHF-style preference checking. Documented recurring issue patterns to inform better labeling and feedback loops for future responses. • Performed quality and accuracy checks on written responses • Used comparative judgments analogous to A/B or preference evaluation • Assessed and prioritized ticket categories based on effectiveness • Produced structured feedback and improvement suggestions

2020 - Present

Education

Work History