Precious Tochi O.

Key Skills

Software

Clickworker

Deep Systems

Micro1

Mercor

OneForma

Telus

Top Subject Matter

LLM evaluation and RLHF-style preference ranking

Software/product QA with structured issue reporting

Technology/AI & Machine Learning

Top Data Types

Text

Image

Audio

Top Task Types

RLHF

Evaluation/Rating

Transcription

Question Answering

Text Summarization

Text Generation

Object Detection

Segmentation

Classification

Freelancer Overview

With 2+ years of experience across AI chatbot evaluation, RLHF-style preference ranking, and crowdsourced QA platforms including uTest, Testio, and Luel, I bring a proven track record of assessing AI outputs for quality, coherence, hallucinations, and instruction-following — and translating those assessments into structured, actionable feedback. I don't just test AI; I understand how to make it better. What sets me apart is rare: six years of science education combined with hands-on AI evaluation experience. I catch what others miss — factual errors, flawed reasoning, STEM inaccuracies — and communicate findings with the precision of someone who has written curricula and instructional content professionally. Detail-oriented, remote-ready, and results-driven. Education includes Bachelor of Science, University of Nigeria, Nsukka (UNN) (2026). AI-training focus includes data types such as Text and labeling workflows including RLHF, Evaluation, and Rating.

Labeling Experience

Software & Product QA Tester (Freelance — uTest, Testio, Luel & Other Platforms)

OtherText

Conducted software and digital service QA testing across multiple freelance platforms by executing test cycles and verifying expected behaviors. Provided detailed, reproducible bug reports and usability feedback that supported development fixes and product iteration. Maintained high accuracy and consistency ratings across remote, asynchronous assignments. • Wrote structured bug reports with reproduction steps • Performed functional testing and usability evaluation • Assessed product quality and reliability through repeated test cycles • Produced feedback intended for developer resolution and refinement

2024 - Present

AI Chat-bot Evaluator & Response Ranker (Remote — Telegram & Multiple Platforms)

OtherTextRLHF

Evaluated AI chatbot responses by assessing accuracy, coherence, instruction-following, and overall output quality in large conversation sets. Performed head-to-head model comparisons and ranked responses while providing preference-based written justifications aligned with RLHF workflows. Identified and documented failures such as hallucinations, factual errors, off-topic replies, and unsafe outputs for iterative model improvement. • Assessed quality dimensions including reasoning quality and adherence to instructions • Ranked responses using preference decisions and written rationales • Produced structured feedback to guide iterative testing cycles • Logged failure modes and documented error patterns

2024 - Present

Education

Work History