Miles D.

Key Skills

Software

No software listed

Top Subject Matter

AI Training

Evaluation

Data Annotation: There is extensive

expert-level experience in Reinforcement Learning from Human Feedback (RLHF)

Supervised Fine-Tuning (SFT)

Top Data Types

Image

Video

Text

Top Task Types

RLHF

Evaluation/Rating

Text Summarization

Text Generation

Entity (NER) Classification

Bounding Box

Freelancer Overview

I am a high-performing Generalist AI Trainer and Multimodal AI Evaluator with extensive experience improving the intelligence, safety, and accuracy of Large Language Models (LLMs) through expert-level RLHF workflows, adversarial testing, and preference ranking. Throughout my career at platforms like Objectways, Alignerr, DataAnnotation, and Outlier, I have specialized in conducting rigorous fact-checking and reasoning audits on high-fidelity datasets across diverse subjects. By stress-testing prompts and identifying subtle logical hallucinations and model biases, I have successfully driven measurable improvements in model grounding, responsiveness, and factual accuracy. My technical precision is backed by a Bachelor’s degree in English and a strong competencies stack that bridges advanced prompt engineering, Supervised Fine-Tuning (SFT), and creative multimodal evaluation utilizing tools like Google Vertex AI and the Adobe Creative Suite. I excel at synthesizing structured rewrites, drafting complex multi-step reasoning prompts, and writing detailed, evidence-based rationales that strictly adhere to evolving safety and style guidelines. I bring an exceptional level of grammar, critical thinking, and guideline compliance to every dataset I deliver, ensuring the highest standards of data integrity.

Labeling Experience

Education

Work History