Open to opportunities

KELLY DYER

@kellydyer

Message

AI data annotation and model evaluation specialist driving high-quality human feedback.

United States

Message

What I'm looking for

I seek remote U.S. roles where I can apply multimodal annotation, model evaluation, and QC to improve model alignment and data quality while collaborating with cross-functional teams.

I am an AI data annotation and model evaluation specialist with 5+ years improving machine learning systems across multimodal, search, and social platforms.

I excel at large-scale annotation, preference ranking, bias detection, and guideline interpretation, delivering precise, high-throughput corrections used for model training and fine-tuning.

My work includes comparative evaluation, relevance scoring, fact-checking, and QC audits, with documented results such as maintaining 98%+ accuracy on weekly quality metrics during freelance engagements.

I am skilled with remote collaboration tools and proprietary labeling platforms, and I provide actionable feedback to cross-functional teams to improve generative model behavior while maintaining strict confidentiality.

Experience

Work history, roles, and key accomplishments

Current

AI Digital Annotation Expert

Current

Mercor Intelligence

Sep 2025 - Present (3 months)

Train and evaluate multimodal AI models by providing high-quality corrections, preference rankings, and QC audits that improve alignment and generative behavior across text, image, audio, and video.

Preference Ranking Auditing Comparative Judgment Model Evaluation Data Integrity

Current

AI Data Annotator

Current

SDL, Inc. (RWS)

Dec 2024 - Present (1 year)

Perform detailed multimodal annotation and research-style prompt creation, validating model responses for factual accuracy and providing structured feedback to improve dataset quality and adherence to confidentiality.

Annotating Multimodal Labeling Fact Checking Quality Assurance Cross Domain Research

AI Data Annotator

Freelance / Gig Work

Jul 2023 - Oct 2024 (1 year 3 months)

Delivered high-accuracy annotation and search relevance evaluations achieving 98%+ weekly quality metrics, refined generative outputs and ranking through comparative scoring and confidentiality best practices.

Annotations Search Relevance Comparative Scoring Quality Confidentiality Remote Collaboration High Throughput Labeling

Search Quality Rater

Appen

Sep 2020 - May 2023 (2 years 8 months)

Evaluated search results and social media content for relevance and intent, improving ranking accuracy and news feed quality through guideline-driven scoring and contribution to large-scale training datasets.

Search Relevance Quality Assurance