Victor J.

Key Skills

Software

Other

Remotasks

Top Subject Matter

Large Language Models (AI, Education, Technology, Research)

Large Language Models (AI, Language, Safety, Reasoning)

Top Data Types

Text

Top Task Types

Fine-tuning

Prompt + Response Writing (SFT)

Freelancer Overview

AI Trainer – Outlier AI. Brings 12+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Other and Remotasks. Education includes Doctor of Philosophy, University of South Carolina (2025) and Master of Arts, Boston University (2024). AI-training focus includes data types such as Text and labeling workflows including Fine-tuning and Prompt + Response Writing (SFT).

Labeling Experience

AI Trainer – Outlier AI

OtherTextFine-tuning

As an AI Trainer at Outlier AI, I trained and evaluated large language models across multiple domains including technology, education, and research. I contributed to the development of prompts, rubrics, and structured feedback systems to improve model quality. I documented inconsistencies and worked collaboratively to refine methods for model advancement. • Enhanced LLM accuracy, coherence, and relevance using structured feedback loops. • Developed and applied rubrics for consistent evaluation of model outputs. • Identified and documented edge cases and model gaps for iterative improvements. • Collaborated with distributed teams to standardize training methodologies.

2024 - Present

Prompt Engineer – Remotasks

RemotasksTextPrompt + Response Writing (SFT)

In my role as Prompt Engineer at Remotasks, I designed, tested, and refined text prompts to optimize and train LLMs. I created and curated diverse prompt-response datasets with classification metadata to enhance model learning. I regularly evaluated AI outputs using structured ranking systems and contributed to iterative improvement strategies. • Developed complex and diverse prompt sets for model accuracy and safety. • Evaluated LLM performance through A/B rankings and structured scoring. • Curated and classified data to support supervised fine-tuning initiatives. • Investigated and iterated on prompt strategies to address model bias and failure modes.

2023 - 2024

Education

Work History