Claire L.

Key Skills

Software

No software listed

Top Subject Matter

AI safety evaluation and LLM failure-mode analysis

AI red team associate

General AI evaluation

Top Data Types

Text

3D Sensor

Document

Top Task Types

Red Teaming

Entity (NER) Classification

Classification

Freelancer Overview

Contract AI Safety and Evaluation Analyst: evaluated LLM outputs for safety, accuracy, and reliability. Brings 14+ years of professional experience across legal operations, contract review, compliance, and structured analysis. Core strengths include Garak and Python. Education includes Associate of Applied Science, Lone Star College System (2022) and Bachelor of Arts, University of Texas at Dallas (2014). AI-training focus includes data types such as Text and 3D Sensor and labeling workflows including Evaluation, Rating, and Red Teaming.

Labeling Experience

AI Safety and Evaluation Analyst (Contract) - Dataannotation

TextRed TeamingEntity (NER) Classification

Evaluated large language model outputs for accuracy, safety, and reliability while identifying and documenting key failure modes. Conducted targeted research and fact verification to reduce misinformation risk in high-volume evaluation workflows. Applied structured evaluation frameworks and judgment-based criteria to assess ambiguous and edge-case scenarios and support consistent safety alignment. • Reviewed model responses for hallucinations, prompt-injection vulnerabilities, and unsafe content • Developed adversarial prompt strategies to expose weaknesses in model behavior • Documented evaluation decisions and escalated high-risk findings for improved model performance

2025 - Present

Contract AI Safety and Evaluation Analyst (DataAnnotation): evaluated LLM outputs for safety, accuracy, and reliability

Text

Contract AI Safety and Evaluation Analyst evaluating LLM outputs for accuracy, safety, and reliability using structured judgment criteria. Identified and documented failure modes such as hallucinations, prompt injection vulnerabilities, and unsafe responses. Performed targeted research and fact verification to detect misinformation and reduce risk in high-volume evaluation workflows. • Applied structured evaluation frameworks and judgment-based criteria to assess ambiguous and edge-case scenarios. • Designed adversarial prompt strategies to systematically expose weaknesses in model behavior. • Documented evaluation decisions and escalated high-risk findings to improve model performance and output quality. • Conducted compliance/safety analysis including safety-filter degradation and contradiction-based drift detection.

2025 - Present

Contract 3D Modeler and Graphic Designer - New York Air Brake

3D SensorEntity (NER) Classification

Created and annotated 3D models for simulation contexts by aligning assets to real-world locations. Worked with geo-spatial simulation data to ensure spatial accuracy of graphical outputs. Used modeling and annotation skills to prepare simulation-ready deliverables for downstream use. • Annotated geo-spatial simulation data for correctness • Aligned 3D assets with real-world locations • Produced graphics and modeling deliverables for simulation needs • Supported integration of 3D content with spatial reference data

2015 - 2015

Education

Work History