Detail-oriented AI training and data annotation specialist with experience creating SFT responses, ranking LLM outputs, auditing factual accuracy, and red teaming for safety. Background also includes search relevance, NER, audio transcription, and team-based QA support, with work shaped by clear guidelines, research, and high accuracy standards.
Overview
1
1
Certification
6
6
years of professional experience
Work History
Generalist & Data Annotation Specialist
Outlier AI / Scale AI
03.2023 - Current
Remote Contractor - San Francisco, CA (Remote) | March 2023 - Present
Created 600+ gold-standard SFT ideal responses for complex prompts in coding, math reasoning, and general knowledge including chain-of-thought explanations.
Evaluated over 2,000 pairs of LLM responses for preference ranking using Helpfulness, Honesty, and Harmlessness criteria.
Justified rankings with 250+ word explanations, achieving 98% approval rating on Outlier leaderboard.
Audited 1,200+ model outputs for factuality and hallucinations, utilizing web research and authoritative sources, and corrected answers with citations, decreasing factual errors by 18%.
Conducted red teaming and safety testing by developing adversarial prompts to evaluate model compliance with disallowed content policies, addressing hate and misinformation.
Executed tasks following Outlier generalist guidelines, including coding helpfulness, factuality check, summarization, and instruction following.
Founder & QA Lead
Handshake, Remotasks, and Outlier-type tasks
Arlington
01.2021 - Current
Founded and led team of 25+ remote taskers for US AI training projects, enhancing operational support for Handshake, Remotasks, and Outlier-type tasks.
Created internal QA system and SOPs, increasing client acceptance rate from 88% to 98.5% and decreasing error rate from 12% to 3%.
Developed workflows for remote tasks in Handshake, Remotasks, and Outlier projects. across Handshake marketplace, Remotasks platform, and Outlier projects.
Managed AI data labeling, prompt evaluation, and content review assignments.
Managed task queues by providing clear instructions and quality guidelines. using clear instructions and quality guidelines.
Team Lead
COLINS TASKING EMPIRE
REMOTE, USA
01.2025 - 01.2026
AI Data Rater
Appen & Invisible Technologies
REMOTE, USA
02.2024 - 02.2025
Data Annotation Specialist
TELUS Digital (TELUS International AI)
06.2022 - 02.2023
Remote - US AI Community | June 2022 - February 2023
Labeled 4,500+ search queries for intent classification as Informational, Transactional, or Navigational, achieving 97% QA score for search result relevance.
Annotated 3,000+ texts for Named Entity Recognition and sentiment analysis, enhancing model accuracy for PERSON, ORGANIZATION, LOCATION, and sentiment categories.
Transcribed and quality-checked 800 hours of US English audio data for Automatic Speech Recognition model, ensuring less than 2% Word Error Rate.
Remote Contractor
Kirkland, WA (Remote)
Kirkland
01.2021 - 05.2022
Evaluated 2,500+ LLM outputs for policy compliance, ensuring adherence to client guidelines on hate speech, self-harm, and misinformation.
Created concise summaries for long-form documents and rewrote AI responses, improving clarity and grammatical accuracy in US English.
Labeled 5,000+ images with bounding boxes and polygon segmentation for object detection in autonomous vehicle dataset, enhancing data quality for training models using CVAT and SuperAnnotate.
Education
Associate of Science - Information Systems - Data Analytics Concentration
University of Texas at Arlington (UTA)
Arlington, TX
05-2021
Some College (No Degree) - Computer Support and Data Management
Austin Community College
Fort Worth, TX
Skills
AI Training & Evaluation: RLHF, RLIAF, Supervised Fine-Tuning (SFT), LLM Evaluation, Prompt Engineering, Chain-of-Thought Reasoning, Preference Ranking, Hallucination Detection, HHH framework, Instruction Following, Red Teaming, AI Safety Alignment
Data Annotation: Text Classification, Named Entity Recognition (NER), Sentiment Analysis, Intent Classification, Search Relevance Rating, Bounding Box, Image Segmentation, Audio Transcription, Content Moderation, Data quality assurance
Tools & Platforms: Labelbox, Label Studio, CVAT, SuperAnnotate, Scale Rapid, Appen Crowd Platform, TELUS Digital Workspace, Jira, Notion, Google Workspace, Microsoft Excel, Python Basics
Prompt evaluation
Model auditing
Data Quality Assurance
Generalist Skills: Research skills, Technical Writing, Summarization, Creative Writing, STEM Reasoning, Critical Thinking, US English Grammar and Style
Strengths: Fast Learner, High Attention to Detail, Meets Tight Deadlines, Written communication
Data Quality Assurance
Personal Information
Availability: 8 AM - 8 PM EST / CST, Weekends Available
Work Permit: US Citizen - Authorized to Work in the United States
Certification
Labelbox Academy - Advanced Data Annotation and QA - 2024
AI Training Specialist and Technical Evaluator at Freelance / Online AI ProjectsAI Training Specialist and Technical Evaluator at Freelance / Online AI Projects