Iwuagwu N.

Key Skills

Software

No software listed

Top Subject Matter

LLM response evaluation for coding assistance

Retrieval-augmented generation evaluation (PDF Q&A)

Structured data quality checks for sales tracking

Top Data Types

Document

Top Task Types

Data Collection

Freelancer Overview

Freelance Tech Support & Data Projects (AI output evaluation and tutoring). Brings 5+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include ClaudeAPI, OpenAI APIs, and GoogleSheets. Education includes Bachelor of Science, University of Nigeria (2021). AI-training focus includes data types such as Computer Code, Programming, and Document and labeling workflows including Evaluation, Rating, and Data Collection.

Labeling Experience

Freelance Tech Support & Data Projects (AI output evaluation and tutoring)

Evaluated AI-generated technical outputs for student users by checking code correctness and identifying hallucinations. Performed quality review using structured judgment to assess whether responses matched expected logic and requirements. Documented observed failure modes and used findings to improve prompt and evaluation approaches for future outputs. • Reviewed AI tool outputs for 30+ student users. • Flagged hallucinations and incorrect code in about 40% of test cases. • Wrote step-by-step Python and SQL explanations to support debugging and learning. • Maintained timely delivery while managing multiple clients with varying requirements.

2022 - Present

WhatsAppSalesTracker — Data Quality Project

DocumentData Collection

Improved dataset quality for a sales tracking workflow by adding input validation rules to catch user entry errors. Cleaned and reviewed raw rows of messy sales data to ensure better consistency and usability for downstream analysis. Reduced bad data through proactive validation before records were stored or processed. • Designed input validation to prevent user errors before data entry. • Reduced bad data by 80% through validation checks. • Reviewed and cleaned 500+ rows of messy sales data for a local business. • Prepared data in Google Sheets-compatible formats for tracking.

2024 - 2024

StudyPDFChatbot — LLM Evaluation Project

Document

Built and evaluated an LLM-based PDF study chatbot by systematically scoring response quality against source PDFs. Applied a rubric to rate factual accuracy, citation correctness, and clarity for each generated answer. Analyzed recurring failure patterns and iterated prompts to reduce hallucinations during testing. • Tested 200+ chatbot responses for factual accuracy vs. source PDFs. • Created a 3-point rubric for correctness, citation accuracy, and clarity. • Logged failure patterns and rewrote prompts to reduce hallucinations by 60% in testing. • Delivered an evaluative workflow using ClaudeAPI and Streamlit for rating.

2024 - 2024

Education

Work History