Austine Wren

Key Skills

Software

Surge AI

AWS SageMaker

VoTT

V7 Labs

Trilldata Technologies

Telus

Toloka

Tagtog

Supervisely

SuperAnnotate

Scale AI

Sloth

Snorkel AI

Remotasks

Redbrick AI

Prodigy

Playment

OpenCV AI Kit (OAK)

OneForma

Mighty AI

Micro1

Mercor

Lionbridge

LightTag

Label Studio

Labelbox

Kili Technology

Img Lab

iMerit

Humanatic

HiveMind

Hasty

Google Cloud Vertex AI

Figure Eight

Encord

Doccano

Diffgram

Deep Systems

Dataturk

Datature

Datasaur

Datumbox

Datatroniq

Dataloop

Data Annotation Tech

CrowdFlower

Anno-Mage

Appen

Argilla

Clickworker

CloudFactory

CrowdSource

CVAT

Top Subject Matter

Drug Discovery & Development

AI in Medical Imaging & Diagnostics

Finance and Insurance

Top Data Types

Text

Image

Document

Top Task Types

Classification

Entity Ner Classification

Object Detection

Cuboid

Polyline

Segmentation

Prompt Response Writing SFT

Data Collection

Function Calling

Evaluation Rating

Transcription

Fine Tuning

Text Summarization

Text Generation

Question Answering

RLHF

Red Teaming

Point Key Point

Bounding Box

Computer Programming Coding

Freelancer Overview

Expert AI Data Annotator specializing in audio and speech synthesis datasets. With a background in Computer Science (M.Sc.) and a strong interest in music technology, I bring 6+ years of experience in structured data analysis. I have experience working with phoneme-level segmentation, time-stamped transcriptions, and pitch/musical note annotation. I am familiar with open-source singing synthesis projects (e.g., GTSinger) and capable of handling the high precision required (millisecond accuracy) for building high-quality singing voice corpora.

Labeling Experience

Freelance AI Data Annotator & Evaluator

AppenText

As a freelance AI data annotator and evaluator, I partnered with top agencies to perform daily tasks across text, image, and video data annotation. I analyzed search queries, rated search engine results, and directly contributed to AI model training for educational and instructional content. I ensured data quality by adhering to rigorous guidelines and consistently achieved audit quality above 95%. • Performed prompt evaluation, content relevance scoring, and sentiment analysis for text data. • Annotated multiple data types, focusing on US educational and instructional domains. • Managed varying project requirements with accuracy and efficiency. • Worked with leading agencies, including RWS, TELUS International, and Appen.

As a freelance AI data annotator and evaluator, I partnered with top agencies to perform daily tasks across text, image, and video data annotation. I analyzed search queries, rated search engine results, and directly contributed to AI model training for educational and instructional content. I ensured data quality by adhering to rigorous guidelines and consistently achieved audit quality above 95%. • Performed prompt evaluation, content relevance scoring, and sentiment analysis for text data. • Annotated multiple data types, focusing on US educational and instructional domains. • Managed varying project requirements with accuracy and efficiency. • Worked with leading agencies, including RWS, TELUS International, and Appen.

2022 - Present

Freelance AI Data Annotator & Evaluator (Audio & Speech Focus)

AudioSegmentation

Contributed to the creation of a high‑quality English singing voice synthesis (SVS) corpus totaling 12 hours of recorded singing. Performed phoneme‑level segmentation with millisecond‑precision timestamps, ensuring accurate alignment of vowel and consonant boundaries for natural speech synthesis. Additionally annotated pitch and musical note information (MIDI note numbers and cents deviation) using reference audio and spectrogram analysis. Followed best practices from open‑source projects like GTSinger to maintain consistency across annotators. Utilized Label Studio for audio waveform labeling and Praat for phoneme boundary verification. Adhered to strict quality checkpoints: 100% of annotations were reviewed by a secondary annotator, and inter‑annotator agreement (IAA) was maintained above 95% for phoneme boundaries and pitch accuracy.

Contributed to the creation of a high‑quality English singing voice synthesis (SVS) corpus totaling 12 hours of recorded singing. Performed phoneme‑level segmentation with millisecond‑precision timestamps, ensuring accurate alignment of vowel and consonant boundaries for natural speech synthesis. Additionally annotated pitch and musical note information (MIDI note numbers and cents deviation) using reference audio and spectrogram analysis. Followed best practices from open‑source projects like GTSinger to maintain consistency across annotators. Utilized Label Studio for audio waveform labeling and Praat for phoneme boundary verification. Adhered to strict quality checkpoints: 100% of annotations were reviewed by a secondary annotator, and inter‑annotator agreement (IAA) was maintained above 95% for phoneme boundaries and pitch accuracy.

2025 - 2025

Vibecode Specialist – Web Scraping & Data Extraction Pipeline

TextEvaluation Rating

I serve as a Vibecode Specialist focused on end-to-end web scraping and data extraction workflows for AI training and e-commerce intelligence applications. Scope: This project involves building reliable, automated data extraction pipelines from complex, dynamic websites. Targets include e-commerce platforms with JavaScript-rendered content, infinite scroll, AJAX-loaded elements, and hierarchical data structures (regions → companies → detailed product listings). Tasks Performed: Developed Python scripts using BeautifulSoup, Selenium, and Requests to extract structured data from multi-level sites. Handled dynamic content by implementing browser automation to wait for JavaScript execution and capture fully rendered pages. Navigated complex site architectures with pagination, login-protected sections, and API endpoints. Cleaned, normalized, and validated raw HTML data, transforming it into structured formats (CSV, JSON, Google Sheets) with consistent presentation. Implemented proxy rotation and error handling to bypass anti-scraping measures and ensure reliable delivery. Leveraged LLMs to accelerate script development, troubleshoot extraction logic, and generate edge-case test scenarios.

I serve as a Vibecode Specialist focused on end-to-end web scraping and data extraction workflows for AI training and e-commerce intelligence applications. Scope: This project involves building reliable, automated data extraction pipelines from complex, dynamic websites. Targets include e-commerce platforms with JavaScript-rendered content, infinite scroll, AJAX-loaded elements, and hierarchical data structures (regions → companies → detailed product listings). Tasks Performed: Developed Python scripts using BeautifulSoup, Selenium, and Requests to extract structured data from multi-level sites. Handled dynamic content by implementing browser automation to wait for JavaScript execution and capture fully rendered pages. Navigated complex site architectures with pagination, login-protected sections, and API endpoints. Cleaned, normalized, and validated raw HTML data, transforming it into structured formats (CSV, JSON, Google Sheets) with consistent presentation. Implemented proxy rotation and error handling to bypass anti-scraping measures and ensure reliable delivery. Leveraged LLMs to accelerate script development, troubleshoot extraction logic, and generate edge-case test scenarios.

2023 - 2024

Education

Work History