Anshu Y.

Key Skills

Software

CVAT

iMerit

Labelbox

Label Studio

Top Subject Matter

No subject matter listed

Top Data Types

Document

Geospatial Tiled Imagery

Image

Text

Top Task Types

Bounding Box

Classification

Data Collection

Entity (NER) Classification

Freelancer Overview

I am a results-driven data annotation specialist with over five years of experience delivering high-quality labeled datasets for AI and machine learning projects. My expertise spans computer vision and NLP domains, where I have led large teams to execute complex annotation tasks using tools like Labelbox, CVAT, Label Studio, and VGG Annotator. I am skilled in a wide range of annotation techniques, including bounding boxes, segmentation, keypoints, and sentiment labeling, and have consistently maintained high accuracy through rigorous quality control and process optimization. My background also includes market and competitor analysis to support project strategy, as well as close collaboration with ML engineers to refine labeling logic and address edge cases. I am passionate about driving team performance, streamlining workflows, and ensuring that every dataset meets the highest standards for AI training.

Labeling Experience

5+ years of data annotations experience

CVATTextBounding BoxEntity (NER) Classification

In this project, I worked extensively on invoice data annotation and document classification, which involved categorizing various financial documents such as Purchase Orders (POs), Receipts, Statements, and Invoices. The goal was to create high-quality labeled datasets to train and evaluate AI models for document understanding and information extraction. After model training, we performed extraction evaluations to verify the model’s ability to accurately capture key entities like invoice numbers, vendor details, item descriptions, quantities, and amounts. Additionally, the AI model performed automated invoice reconciliation by matching extracted invoice data with corresponding Purchase Orders to validate data accuracy and consistency. This project required a strong focus on data accuracy, financial domain understanding, and annotation quality control, along with close collaboration between annotation and model evaluation teams.

2021

Education

Work History