Chinedu Alozie

Key Skills

Software

Label Studio

Top Subject Matter

General LLM/AI

Finance/Wealth Management

Security/App User Access

Top Data Types

Text

Document

Top Task Types

Prompt + Response Writing (SFT)

Classification

Freelancer Overview

AI Tool Workflow Analyst. Brings 7+ years of professional experience across legal operations, contract review, compliance, and structured analysis. Core strengths include Label Studio and Google Workspace. Education includes Bachelor of Science, University of Nigeria, Nsukka (2021). AI-training focus includes data types such as Text and labeling workflows including Evaluation, Rating, and Prompt + Response Writing (SFT).

Labeling Experience

Freelance Copywriter – Hydrapolis Capital

TextPrompt Response Writing SFT

As a freelance copywriter for finance-domain SFT data, I created policy-compliant prompts and ideal responses. I generated A/B response variants and labeled preferred selections for clarity and factuality. I also reviewed outputs for policy violations in wealth management scenarios. • Produced SFT training data centered on financial compliance. • Labeled and preferred responses for factual correctness. • Evaluated model outputs for finance policies. • Utilized structured prompting techniques.

As a freelance copywriter for finance-domain SFT data, I created policy-compliant prompts and ideal responses. I generated A/B response variants and labeled preferred selections for clarity and factuality. I also reviewed outputs for policy violations in wealth management scenarios. • Produced SFT training data centered on financial compliance. • Labeled and preferred responses for factual correctness. • Evaluated model outputs for finance policies. • Utilized structured prompting techniques.

2025 - 2025

Data Annotation Specialist – Estate Security App Project

Label StudioTextClassification

For the Estate Security App Project, I designed and executed annotation tasks focused on classifying user intents in security-related prompts. I labeled complex, edge-case access-control requests and wrote gold-standard responses for ambiguous user prompts. I also performed high-accuracy spot QA on peer annotations to ensure label consistency. • Developed intent classification labels for security domain. • Handled ambiguous or edge user scenarios for access control. • Created QA metrics and guidelines for annotation quality. • Improved peer labeling accuracy through guideline adherence.

For the Estate Security App Project, I designed and executed annotation tasks focused on classifying user intents in security-related prompts. I labeled complex, edge-case access-control requests and wrote gold-standard responses for ambiguous user prompts. I also performed high-accuracy spot QA on peer annotations to ensure label consistency. • Developed intent classification labels for security domain. • Handled ambiguous or edge user scenarios for access control. • Created QA metrics and guidelines for annotation quality. • Improved peer labeling accuracy through guideline adherence.

2024 - 2025

AI Tool Workflow Analyst

Label StudioText

As an AI Tool Workflow Analyst, I evaluated over 1,000 LLM-generated responses for accuracy, helpfulness, and safety. I labeled and ranked AI outputs using detailed rubrics and flagged hallucinations, bias, and policy issues. I authored ideal prompt responses and documented common failure modes to improve LLM model guidance. • Conducted systematic evaluation of LLM outputs on ChatGPT, Claude, and Gemini. • Applied rubric-based labels for response quality and compliance. • Wrote model-improving responses for SFT datasets. • Reported annotation insights to update guidelines.

As an AI Tool Workflow Analyst, I evaluated over 1,000 LLM-generated responses for accuracy, helpfulness, and safety. I labeled and ranked AI outputs using detailed rubrics and flagged hallucinations, bias, and policy issues. I authored ideal prompt responses and documented common failure modes to improve LLM model guidance. • Conducted systematic evaluation of LLM outputs on ChatGPT, Claude, and Gemini. • Applied rubric-based labels for response quality and compliance. • Wrote model-improving responses for SFT datasets. • Reported annotation insights to update guidelines.

2024 - 2025

Education

Work History