For employers

Hire this AI Trainer

Sign in or create an account to invite AI Trainers to your job.

LoginCreate account
Invite to Job
P
Paul Odantabao

Paul Odantabao

AI Data Annotator & Human-in-the-Loop Reviewer (Invoice Tidy Project)

Philippines flagN/A, Philippines
$15.00/hrEntry LevelAxiom AIGoogle Cloud Vertex AI

Key Skills

Software

Axiom AI
Google Cloud Vertex AIGoogle Cloud Vertex AI

Top Subject Matter

Data Entry, Annotation & Extraction
Bookkeeping/Financial Document Processing
General AI / Chatbot Evaluation

Top Data Types

DocumentDocument
ImageImage

Top Task Types

ClassificationClassification
PolygonPolygon
Bounding BoxBounding Box
Computer Programming/CodingComputer Programming/Coding

Freelancer Overview

My experience centers on architecting Human-in-the-Loop (HITL) workflows and evaluating complex AI outputs, most notably through the development of Invoice Tidy. By engineering an automated document-processing tool with a dynamic confidence-scoring algorithm, I designed the exact exception-first review pipelines used in industry-standard AI data labeling. This hands-on work with OCR performance, document classification, and structured data extraction ensures I am highly calibrated to evaluate, verify, and correct automated outputs with a strict focus on data fidelity. What sets me apart for advanced AI training is my rigorous mathematical foundation and coding expertise. Beyond standard data annotation, my background in electrical engineering and deep learning—highlighted by my research developing DeepAR probabilistic forecasting pipelines—allows me to tackle complex logic and algorithmic assessments. As a developer proficient in C, CUDA, and Python, and a daily power user of agentic tools like Cursor and Claude, I am uniquely positioned to identify LLM logic failures, optimize code generation, and provide expert-level technical feedback.

Entry LevelEnglish

Labeling Experience

Data Extraction and Human-in-the-Loop AI Reviewer (Invoice Tidy Project)

DocumentClassification
The candidate designed an AI-powered document extraction and review engine focused on invoice data. This involved building a system that parses unstructured invoice PDFs, extracts key fields, and routes uncertain outputs for human-in-the-loop verification. The workflow included optimizing a UI for efficient manual review and flagging low-confidence data. • Designed extraction pipelines for structured data. • Implemented a confidence-scoring algorithm for data quality control. • Developed human-in-the-loop verification workflows. • Ensured clean output for financial bookkeeping import.

The candidate designed an AI-powered document extraction and review engine focused on invoice data. This involved building a system that parses unstructured invoice PDFs, extracts key fields, and routes uncertain outputs for human-in-the-loop verification. The workflow included optimizing a UI for efficient manual review and flagging low-confidence data. • Designed extraction pipelines for structured data. • Implemented a confidence-scoring algorithm for data quality control. • Developed human-in-the-loop verification workflows. • Ensured clean output for financial bookkeeping import.

2023 - Present

Education

P

Philippine Science High School - Zamboanga Peninsula Region Campus

High School Diploma, Science

High School Diploma
2016 - 2022
M

Mindanao State University - Iligan Institute of Technology

Bachelor of Science, Electrical Engineering

Bachelor of Science
2022

Work History

N

Nua

Machine Learning Engineer

N/A
2025 - Present
I

Invoice Tidy

Software Engineer

N/A
2023 - 2024