Jonathan S.

Key Skills

Software

No software listed

Top Subject Matter

Healthcare-oriented generative AI & multimodal model behavior

Multimodal visual reasoning & preference-based evaluation

Multimodal evaluation (LLMs and image edits) for vision-language tasks

Top Data Types

Audio

Text

Image

Document

Top Task Types

Prompt + Response Writing (SFT)

Question Answering

Fine-tuning

Data Collection

Freelancer Overview

Lead AI Specialist, Alignerr AI (Nov 2025 - Present). Brings 5+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Internal and Proprietary Tooling. Education includes Doctor of Philosophy, Rochester Institute of Technology (2026) and Master of Science, Really Great University (2022). AI-training focus includes data types such as Audio, Text, and Image and labeling workflows including Prompt + Response Writing (SFT), Question Answering, and Evaluation.

Labeling Experience

Lead AI Specialist, Alignerr AI (Nov 2025 - Present)

AudioPrompt + Response Writing (SFT)

Led prompt-centric AI training and evaluation efforts, focusing on improving model behavior across conversational turns. Produced challenging scenarios intended to surface weaknesses and drive iterative improvements through multi-turn interactions. Supported accuracy gains across multimodal tasks including audio. • Designed and authored complex, high-quality prompts • Created multi-turn scenarios to expose model weaknesses • Guided improvements using observed failures • Contributed to accuracy enhancements across audio, visual, and text tasks

2025 - Present

AI Specialist, Litero AI (June 2025 - Sept 2025)

TextQuestion Answering

Trained and evaluated a model’s visual reasoning capabilities by creating and using multimodal question-answer data. Performed structured response evaluation using multiple quality dimensions and preference ranking aligned to prompt intent. Compared model outputs using rubric-based criteria to determine which responses demonstrated stronger reasoning and communication. • Created challenging multimodal Q&A pairs for visual reasoning • Evaluated responses across multiple quality dimensions • Reviewed and compared outputs using structured criteria • Assigned preference rankings based on prompt intent

2025 - 2025

AI/ Machine learning Researcher, Scale AI / Outlier (June 2024 - May 2025)

Image

Conducted LLM and multimodal evaluation by selecting better responses using a project scoring scale and writing justifications for clarity. Benchmarked and graded model performance for chart and diagram question answering using verified visual reasoning questions. Performed blind evaluations of LLMs and compared models by executing complex tasks, ranking user experience, and documenting detailed qualitative rationales. • Selected preferred model responses using a scoring scale • Authored response justifications and global qualitative rationales • Evaluated AI-generated image edits across multiple dimensions to identify failure patterns • Designed and benchmarked chart/diagram visual question answering tasks with verified answers

2024 - 2025

Machine Learning Scientist, Arizona AI Research Institute (Nov 2022 - Feb 2024)

DocumentFine-tuning

Built LLM evaluation pipelines and improved annotation consistency through systematic evaluation and processing. Authored scaled data labeling systems intended for use with high-level reasoners. Trained data via statistical filtering and preprocessing to support downstream model training and quality improvements. • Created pipelines for LLM evaluation with reduced annotation inconsistencies • Authored scalable labeling systems for reasoning models • Applied statistical filtering and preprocessing to training data • Supported improved consistency through evaluation-driven workflows

2022 - 2024

Data Labeling Manager, Phoenix Data Works (June 2021 - Feb 2022)

DocumentData Collection

Managed large-scale annotation efforts by planning and controlling teams of expert annotators across multiple projects. Ensured quality in labeled corpora by teaching statistical quality control practices to simplify annotator work. Achieved high accuracy rates on large volumes of labeled data used for training. • Planned and coordinated 50+ expert annotators across 5-8 projects • Implemented workflow controls for annotation production • Taught statistical quality control methods for consistent labeling • Delivered 95% accuracy in large-scale labeled corpora

2021 - 2022

Education

Work History