Jaylen Harris

Key Skills

Software

Mercor

Telus

Remotasks

Scale AI

Top Subject Matter

Stem Domain Expertise

Academic Evaluation

AI Model Assessment

Top Data Types

Text

Video

Audio

Image

Document

Top Task Types

Transcription

Classification

Entity (NER) Classification

Question Answering

Object Detection

RLHF

Evaluation/Rating

Bounding Box

Red Teaming

Prompt + Response Writing (SFT)

Point/Key Point

Freelancer Overview

With over six years of professional experience, I specialize in high-precision research, complex workflow management, and the rigorous evaluation of AI-generated content. My expertise is grounded in a strong academic foundation from McNeese State University, where I earned an Associate of Science in Applied and Natural Science and am currently advancing toward a Bachelor of Science in the same field. Throughout my career, I have become a core contributor to major AI training initiatives at Mercor, Outlier, and Telus International, consistently delivering high-quality data that drives model accuracy and reliability. My technical focus centers on the intersection of STEM academic standards and multi-modal AI training, including text, video, and audio data streams. I possess deep familiarity with diverse labeling workflows, ranging from comparative evaluation and quality rating to precise transcription and data auditing within the Applied and Natural Science domains. By combining scientific literacy with a mastery of human-in-the-loop methodologies, I ensure that complex technical outputs meet the highest standards of logical consistency and subject-matter integrity.

Labeling Experience

STEM - Academic Evaluator Expert

MercorText

This role involved preference ranking and A/B testing to determine the most helpful and contextually appropriate AI model outputs. I analyzed qualitative and quantitative data for trends and ensured model outputs met rigorous academic and industry standards in STEM-based reasoning. Decision-making was informed by compliance with ethical guidelines while assessing high-complexity technical content. • Performed comparative output analysis for logical and contextual alignment. • Used data analysis tools to enhance model selection accuracy. • Maintained confidentiality and quality while handling technical datasets. • Contributed to ongoing evaluation protocols for academic reasoning tasks.

This role involved preference ranking and A/B testing to determine the most helpful and contextually appropriate AI model outputs. I analyzed qualitative and quantitative data for trends and ensured model outputs met rigorous academic and industry standards in STEM-based reasoning. Decision-making was informed by compliance with ethical guidelines while assessing high-complexity technical content. • Performed comparative output analysis for logical and contextual alignment. • Used data analysis tools to enhance model selection accuracy. • Maintained confidentiality and quality while handling technical datasets. • Contributed to ongoing evaluation protocols for academic reasoning tasks.

2026 - Present

AI Audio Evaluation Specialist

Audio

I evaluated model-generated audio for human-like speech, prosody, and tonal alignment across diverse scenarios. My duties included scoring for narrative consistency, logical reasoning, and contextual relevance in audio-based AI tasks. I delivered structured feedback to improve reliability and flow in AI conversational outputs. • Judged intent detection and instructional alignment in audio. • Applied advanced rubrics to linguistic and emotional markers. • Synthesized multimodal data for improved content categorization. • Identified and reported model failure modes for system improvement.

I evaluated model-generated audio for human-like speech, prosody, and tonal alignment across diverse scenarios. My duties included scoring for narrative consistency, logical reasoning, and contextual relevance in audio-based AI tasks. I delivered structured feedback to improve reliability and flow in AI conversational outputs. • Judged intent detection and instructional alignment in audio. • Applied advanced rubrics to linguistic and emotional markers. • Synthesized multimodal data for improved content categorization. • Identified and reported model failure modes for system improvement.

2026 - 2026

AI Video Evaluation & Content Quality Specialist

Video

I performed video-based evaluation of multimodal AI outputs, focusing on assessment of model reasoning, factual alignment, and logical consistency. Detailed annotation involved categorizing video segments with precise timestamps and thorough summaries. Complex visual data analysis ensured accurate model response to nuanced action and intent. • Conducted A/B testing for helpfulness and contextual appropriateness. • Labeled and summarized intricate video clips for model refinement. • Contributed to edge-case analysis and documentation of reasoning gaps. • Strengthened the robustness of multimodal foundational models.

I performed video-based evaluation of multimodal AI outputs, focusing on assessment of model reasoning, factual alignment, and logical consistency. Detailed annotation involved categorizing video segments with precise timestamps and thorough summaries. Complex visual data analysis ensured accurate model response to nuanced action and intent. • Conducted A/B testing for helpfulness and contextual appropriateness. • Labeled and summarized intricate video clips for model refinement. • Contributed to edge-case analysis and documentation of reasoning gaps. • Strengthened the robustness of multimodal foundational models.

2026 - 2026

AI Evaluation & Data Annotation Specialist

Text

I conducted rigorous evaluation and annotation of AI-generated text outputs across diverse domains, with an emphasis on instructional alignment and logical structure. Tasks included Chain-of-Thought verification, red teaming, and creation of Golden Response datasets for supervised fine-tuning. Adapted quickly to new rubrics and workflows, balancing creativity and technical problem-solving. • Implemented adversarial stress-testing for model robustness. • Generated SFT reference datasets for benchmarking. • Evaluated multi-turn reasoning and logic. • Navigated dynamic project demands and annotation guidelines.

I conducted rigorous evaluation and annotation of AI-generated text outputs across diverse domains, with an emphasis on instructional alignment and logical structure. Tasks included Chain-of-Thought verification, red teaming, and creation of Golden Response datasets for supervised fine-tuning. Adapted quickly to new rubrics and workflows, balancing creativity and technical problem-solving. • Implemented adversarial stress-testing for model robustness. • Generated SFT reference datasets for benchmarking. • Evaluated multi-turn reasoning and logic. • Navigated dynamic project demands and annotation guidelines.

2025 - 2026

Audio Data Collection & Transcription

TelusVideoTranscriptionClassification

I ensured quality outcomes by performing detailed annotation and captioning of video clips with precise timestamps and summaries. I brought structured thinking and linguistic clarity to all multimedia annotation tasks to enhance model accessibility. My efforts supported better training data for AI systems relying on multimodal content. • Delivered accurate summaries for complex video segments. • Managed multilayered annotation projects with attention to detail. • Enhanced accessibility through meticulous captioning practices. • Contributed to improved model understanding of video content.

I ensured quality outcomes by performing detailed annotation and captioning of video clips with precise timestamps and summaries. I brought structured thinking and linguistic clarity to all multimedia annotation tasks to enhance model accessibility. My efforts supported better training data for AI systems relying on multimodal content. • Delivered accurate summaries for complex video segments. • Managed multilayered annotation projects with attention to detail. • Enhanced accessibility through meticulous captioning practices. • Contributed to improved model understanding of video content.

2025 - 2025

Education

Work History