Jessica K.

Key Skills

Software

Don't disclose

Top Subject Matter

K-12 educational content and general LLM conversation evaluation

Multimodal AI assessment and RLHF feedback for generative models

Top Data Types

Text

Image

Top Task Types

Prompt + Response Writing (SFT)

Data Collection

Evaluation/Rating

Transcription

Text Summarization

RLHF

Freelancer Overview

AI Conversation Auditor (K-12 Expert) — evaluated and rated AI conversations for quality, accuracy, and safety at Outlie. Core strengths include Don't disclose. Education includes Bachelor of Science, Southern New Hampshire University (2024). AI-training focus includes data types such as Text and Image and labeling workflows including Evaluation and Rating.

Labeling Experience

AI Training Specialist — created/evaluated multimodal training data and provided rated feedback for RLHF at Outlier.ai.

Don't discloseImage

Created and evaluated multimodal AI training data involving interactions across text, images, audio, and video. Audited and rated AI outputs for quality, correctness, and compliance with detailed project guidelines. Contributed human feedback to reinforcement learning workflows to refine generative AI system performance.• Performed multimodal response evaluation across image-related interactions as part of training data review.• Applied annotation guidelines to ensure consistent, high-quality evaluation outputs.• Provided reasoning-focused and instruction-following feedback for improved model behavior.• Maintained quality and productivity metrics while adapting to changing evaluation standards.

2024 - 2026

AI Conversation Auditor (K-12 Expert) — evaluated and rated AI conversations for quality, accuracy, and safety at Outlier.ai.

Don't discloseText

Evaluated AI-generated conversations to determine factual accuracy, educational quality, and adherence to project guidelines. Produced structured ratings and feedback based on complex prompt criteria and model response quality. Identified errors, inconsistencies, and safety concerns to improve the usefulness and reliability of generative AI outputs.• Assessed multi-turn dialogues for reasoning, creativity, factual accuracy, and instruction-following.• Applied detailed annotation guidelines to maintain consistent evaluation across tasks.• Delivered structured feedback to enhance model performance and reduce safety issues.• Conducted comparative response ranking to support model improvement cycles.

2023 - 2024

Education

Work History