Matthew Smalarz

Key Skills

Software

Mercor

Other

Telus

Data Annotation Tech

Labelbox

Top Subject Matter

AI Chatbots

General Knowledge

Media Content

Top Data Types

Text

Audio

Document

Top Task Types

Text Generation

Question Answering

Prompt + Response Writing (SFT)

Transcription

Classification

Segmentation

Object Detection

Text Summarization

RLHF

Fine-tuning

Evaluation/Rating

Data Collection

Freelancer Overview

AI Trainer, Subject Matter Expert. Brings 14+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Prolific, Mercor, Turing, and Handshake. Education includes Doctor of Philosophy, University of Rochester (2016) and pursuing a Master of Science in Analytics, Georgia Institute of Technology, starting in 2026. AI-training focus includes data types such as Text and Audio and labeling workflows including Text Generation, Evaluation, and Rating. Subject Matter Expert in History and Political Science, with additional experience as an expert prompt and rubric writer.

Labeling Experience

Media Analysis Expert

MercorText

Applied reasoning and judgment to AI media content outputs to ensure contextual accuracy and identify/correct cultural nuances. Monitored and resolved factual inaccuracies and bias in AI-generated media responses. Collaborated in structuring and enhancing quality in AI-created media content. • Evaluated contextual accuracy of AI-produced media. • Detected and reported bias and factual errors in outputs. • Contributed to AI media contextualization efforts. • Improved cultural relevance and nuance in AI content.

Applied reasoning and judgment to AI media content outputs to ensure contextual accuracy and identify/correct cultural nuances. Monitored and resolved factual inaccuracies and bias in AI-generated media responses. Collaborated in structuring and enhancing quality in AI-created media content. • Evaluated contextual accuracy of AI-produced media. • Detected and reported bias and factual errors in outputs. • Contributed to AI media contextualization efforts. • Improved cultural relevance and nuance in AI content.

2026 - Present

AI Trainer

TextText Generation

Created carefully scaffolded prompts to improve chatbot model responses and edited/reviewed AI generated outputs for quality assurance. Applied critical thinking and reasoning to identify inconsistencies and fallacies to improve response quality. Enhanced AI chatbot performance through active evaluation and revision of model responses. • Developed detailed and structured chat prompts for LLM training. • Identified factual inaccuracies and inconsistencies in language model responses. • Conducted quality assurance reviews of generated outputs. • Improved overall quality and factual accuracy of AI model responses.

Created carefully scaffolded prompts to improve chatbot model responses and edited/reviewed AI generated outputs for quality assurance. Applied critical thinking and reasoning to identify inconsistencies and fallacies to improve response quality. Enhanced AI chatbot performance through active evaluation and revision of model responses. • Developed detailed and structured chat prompts for LLM training. • Identified factual inaccuracies and inconsistencies in language model responses. • Conducted quality assurance reviews of generated outputs. • Improved overall quality and factual accuracy of AI model responses.

2026 - Present

Generative AI History and Writing Subject Matter Expert and Trainer

OtherTextQuestion Answering

Developed sophisticated question and answer pairs on historical topics using large language models. Ensured questions fostered deep, multi-step reasoning and high academic standards. Collaborated with teams to refine training content for both history and writing-focused generative AI models. • Produced detailed questions and answers for LLM training. • Maintained critical standard of analytical reasoning. • Enhanced generative model writing capability. • Collaborated to optimize content refinement.

Developed sophisticated question and answer pairs on historical topics using large language models. Ensured questions fostered deep, multi-step reasoning and high academic standards. Collaborated with teams to refine training content for both history and writing-focused generative AI models. • Produced detailed questions and answers for LLM training. • Maintained critical standard of analytical reasoning. • Enhanced generative model writing capability. • Collaborated to optimize content refinement.

2025 - Present

Generative AI History PhD Data Partner and Data Annotator

TelusTextQuestion Answering

Developed original question-answer pairs on advanced historical topics for generative AI models. Ensured logical, multi-step reasoning and academic rigor in all solutions. Worked with interdisciplinary teams to enhance historical content and reasoning quality for LLMs. • Created high-quality historical Q&A data for LLMs. • Maintained rigorous academic standards in datasets. • Supported enhancement of AI historical accuracy. • Collaborated for continuous dataset improvement.

Developed original question-answer pairs on advanced historical topics for generative AI models. Ensured logical, multi-step reasoning and academic rigor in all solutions. Worked with interdisciplinary teams to enhance historical content and reasoning quality for LLMs. • Created high-quality historical Q&A data for LLMs. • Maintained rigorous academic standards in datasets. • Supported enhancement of AI historical accuracy. • Collaborated for continuous dataset improvement.

2025 - Present

AI Fellow, Handshake MOVE Program

OtherTextPrompt Response Writing SFT

Formulated prompts and creative AI tasks for educational research purposes. Reviewed AI responses for structural alignment and factual accuracy. Collaborated on refining and enhancing model-generated educational content. • Developed extensive prompts for AI educational tasks. • Reviewed outputs for accuracy and alignment. • Supported cross-functional refinement of datasets. • Enhanced critical thinking in AI responses.

Formulated prompts and creative AI tasks for educational research purposes. Reviewed AI responses for structural alignment and factual accuracy. Collaborated on refining and enhancing model-generated educational content. • Developed extensive prompts for AI educational tasks. • Reviewed outputs for accuracy and alignment. • Supported cross-functional refinement of datasets. • Enhanced critical thinking in AI responses.

2025 - Present

Show more

Education

Work History