Elizabeth Kaufman

Key Skills

Software

Scale AI

Appen

Other

Top Subject Matter

LLM Training

Nlp Domain Expertise

Computer Vision

Top Data Types

Text

Image

Document

Top Task Types

Classification

Freelancer Overview

Senior AI Data Annotation Specialist. Brings 7+ years of professional experience across legal operations, contract review, compliance, and structured analysis. Core strengths include Scale AI, Appen, and Other. Education includes Bachelor of Science, University of Michigan, Ann Arbor (2018). AI-training focus includes data types such as Multimodal and Text and labeling workflows including Evaluation, Rating, and Classification.

Labeling Experience

Senior AI Data Annotation Specialist

Scale AI

Acted as the Senior AI Data Annotation Specialist responsible for multi-domain data labeling for LLM pipelines. Labeled and evaluated over 300,000 text, image, and multimodal data points, maintaining a high-quality and accuracy benchmark. Carried out RLHF preference ranking, hallucination audits, and adversarial prompt red teaming for improving model safety and factuality. • Led QA reviews and annotation guideline development • Conducted structured rubric-based error flagging • Participated in red teaming for safety and robustness • Maintained a personal annotation accuracy score of 98.4%

Acted as the Senior AI Data Annotation Specialist responsible for multi-domain data labeling for LLM pipelines. Labeled and evaluated over 300,000 text, image, and multimodal data points, maintaining a high-quality and accuracy benchmark. Carried out RLHF preference ranking, hallucination audits, and adversarial prompt red teaming for improving model safety and factuality. • Led QA reviews and annotation guideline development • Conducted structured rubric-based error flagging • Participated in red teaming for safety and robustness • Maintained a personal annotation accuracy score of 98.4%

2022 - Present

Data Labeling Specialist

AppenClassification

Performed multimodal data labeling tasks for image, audio, point cloud, and text data on projects including autonomous vehicles, ASR, and sentiment analysis models. Labeled bounding boxes, polygons, NER, keypoints, and 3D cuboids for a range of perception and NLP applications. Transcribed, segmented, and classified 500+ hours of multilingual audio and contributed to e-commerce-focused NLP annotation. • Executed 1,200–1,500 annotations daily with <1% error • Ranked top 10% for contributor quality • Labeled 3D cuboids for LiDAR robotics • Used Labelbox and CVAT for accurate data labeling

Performed multimodal data labeling tasks for image, audio, point cloud, and text data on projects including autonomous vehicles, ASR, and sentiment analysis models. Labeled bounding boxes, polygons, NER, keypoints, and 3D cuboids for a range of perception and NLP applications. Transcribed, segmented, and classified 500+ hours of multilingual audio and contributed to e-commerce-focused NLP annotation. • Executed 1,200–1,500 annotations daily with <1% error • Ranked top 10% for contributor quality • Labeled 3D cuboids for LiDAR robotics • Used Labelbox and CVAT for accurate data labeling

2020 - 2021

Freelance AI Trainer & Annotator

OtherTextClassification

Served as a freelance annotator and AI trainer for NLP research and model benchmarking across multiple crowdsourcing platforms. Completed microtask annotations involving text classification, semantic similarity, and toxicity detection for early-stage language model development. Evaluated model-generated creative writing, code samples, and Q&A for supervised fine-tuning datasets and annotated satellite geospatial imagery. • Delivered 50,000+ labeled microtasks • Provided structured feedback for SFT data • Labeled geospatial data for land-use mapping • Worked with Prolific, Remotasks, and Surge AI platforms

Served as a freelance annotator and AI trainer for NLP research and model benchmarking across multiple crowdsourcing platforms. Completed microtask annotations involving text classification, semantic similarity, and toxicity detection for early-stage language model development. Evaluated model-generated creative writing, code samples, and Q&A for supervised fine-tuning datasets and annotated satellite geospatial imagery. • Delivered 50,000+ labeled microtasks • Provided structured feedback for SFT data • Labeled geospatial data for land-use mapping • Worked with Prolific, Remotasks, and Surge AI platforms

2019 - 2020

Education

Work History