Mathew Hayer

Key Skills

Software

Labelbox

CVAT

AWS SageMaker

Supervisely

Other

Roboflow

Scale AI

Top Subject Matter

Trust & Safety

LLM Evaluation

Motion Detection

Top Data Types

Image

Video

Audio

Text

Document

Top Task Types

Bounding Box

Tracking

Transcription

Entity Ner Classification

Classification

Data Collection

Segmentation

Polygon

Freelancer Overview

Trust & Safety / LLM Evaluation Specialist. Core strengths include Scale AI, CVAT, and Labelbox. AI-training focus includes data types such as Text, Video, and Image and labeling workflows including Evaluation, Rating, and Tracking.

Labeling Experience

Trust & Safety / LLM Evaluation Specialist

Scale AIText

Assessed and labeled AI-generated text outputs in Spanish and English for Trust & Safety and LLM evaluation. Conducted adversarial testing, red-teaming, and policy alignment checks to strengthen model robustness and ensure safety compliance. Performed RLHF labeling, content categorization, and output ranking for reinforcement learning from human feedback. • Evaluated text generation quality, correctness, and policy compliance • Applied emotional resilience in sensitive content scenarios • Used structured guidelines for labeling and rating • Supported ongoing model improvement and dataset quality

Assessed and labeled AI-generated text outputs in Spanish and English for Trust & Safety and LLM evaluation. Conducted adversarial testing, red-teaming, and policy alignment checks to strengthen model robustness and ensure safety compliance. Performed RLHF labeling, content categorization, and output ranking for reinforcement learning from human feedback. • Evaluated text generation quality, correctness, and policy compliance • Applied emotional resilience in sensitive content scenarios • Used structured guidelines for labeling and rating • Supported ongoing model improvement and dataset quality

2024 - Present

PDF Forms Data Collection Specialist (Simplified Chinese) – Remote / Freelance

DocumentData Collection

I conduct online research to locate and collect blank PDF forms from official Chinese-language sources. I verify file authenticity and ensure document clarity, completeness, and formatting. I organize, label, and quality check files, following structured dataset guidelines for integrity. • Used Google Search, Baidu, and official websites for document sourcing • Ensured only authentic, publicly accessible forms were included • Performed comprehensive quality control on all files • Maintained dataset integrity and tracking for structured data management

I conduct online research to locate and collect blank PDF forms from official Chinese-language sources. I verify file authenticity and ensure document clarity, completeness, and formatting. I organize, label, and quality check files, following structured dataset guidelines for integrity. • Used Google Search, Baidu, and official websites for document sourcing • Ensured only authentic, publicly accessible forms were included • Performed comprehensive quality control on all files • Maintained dataset integrity and tracking for structured data management

2024 - Present

Singing Voice & Audio Annotation Specialist

CVATVideoTranscriptionClassification

As a Singing Voice & Audio Annotation Specialist, I performed detailed phoneme-level annotation of audio with precise start and end times. I labeled pitch and conducted per-segment frequency analysis for singing voice synthesis datasets. I followed strict guidelines and managed structured workflows for quality delivery. • Used Praat and Librosa for advanced audio analysis and annotation. • Produced Praat TextGrid files adhering to GTSinger-style conventions. • Focused on dataset accuracy, milestone tracking, and quality assurance. • Worked with complex datasets including vibrato and pitch variations.

As a Singing Voice & Audio Annotation Specialist, I performed detailed phoneme-level annotation of audio with precise start and end times. I labeled pitch and conducted per-segment frequency analysis for singing voice synthesis datasets. I followed strict guidelines and managed structured workflows for quality delivery. • Used Praat and Librosa for advanced audio analysis and annotation. • Produced Praat TextGrid files adhering to GTSinger-style conventions. • Focused on dataset accuracy, milestone tracking, and quality assurance. • Worked with complex datasets including vibrato and pitch variations.

2023 - Present

AI Training Specialist

LabelboxImageBounding BoxEntity Ner Classification

As an AI Training Specialist, I annotated diverse datasets containing images, videos, audio, and text to create training data for machine learning models. I applied bounding boxes, segmentation masks, and classification labels to images, with additional object tracking for video projects. My responsibilities extended to extracting emails, validating datasets, and collaborating with teams to maintain annotation quality. • Annotated images and videos with bounding boxes, masks, and tracking information. • Used tools such as Labelbox, CVAT, and AWS SageMaker to conduct labeling workflows. • Ensured rigorous quality assurance and dataset validation prior to AI training. • Contributed to refining labeling guidelines and improving dataset quality.

As an AI Training Specialist, I annotated diverse datasets containing images, videos, audio, and text to create training data for machine learning models. I applied bounding boxes, segmentation masks, and classification labels to images, with additional object tracking for video projects. My responsibilities extended to extracting emails, validating datasets, and collaborating with teams to maintain annotation quality. • Annotated images and videos with bounding boxes, masks, and tracking information. • Used tools such as Labelbox, CVAT, and AWS SageMaker to conduct labeling workflows. • Ensured rigorous quality assurance and dataset validation prior to AI training. • Contributed to refining labeling guidelines and improving dataset quality.

2023 - Present

Data Labeling & Data Extraction Specialist

SuperviselyImageBounding BoxEntity Ner Classification

In text annotation, NLP tagging, and evaluation-labeling roles, I conducted entity recognition, prompt-response evaluation, and categorization for structured and unstructured data. Leveraging platforms like Labelbox, Scale AI, CVAT, and SageMaker Ground Truth, I defined labeling guidelines and ensured data quality through comprehensive validation. My efforts enabled improved AI comprehension and NLP model performance for a variety of language-driven projects. • Tagged named entities and relationship types for NER datasets • Reviewed prompt-response pairs and rated label quality • Established and maintained QA benchmarks for text data • Facilitated dataset enhancements for downstream language model use

In text annotation, NLP tagging, and evaluation-labeling roles, I conducted entity recognition, prompt-response evaluation, and categorization for structured and unstructured data. Leveraging platforms like Labelbox, Scale AI, CVAT, and SageMaker Ground Truth, I defined labeling guidelines and ensured data quality through comprehensive validation. My efforts enabled improved AI comprehension and NLP model performance for a variety of language-driven projects. • Tagged named entities and relationship types for NER datasets • Reviewed prompt-response pairs and rated label quality • Established and maintained QA benchmarks for text data • Facilitated dataset enhancements for downstream language model use

2022 - 2023

Show more

Education

Work History