Mariya Lytvynchuk Muntian

Key Skills

Software

Scale AI

Top Subject Matter

No subject matter listed

Top Data Types

Image

Text

Top Task Types

Bounding Box

Classification

Evaluation Rating

Text Generation

Translation Localization

Freelancer Overview

Over the past six months at Scale AI, I've gained extensive experience in shaping and evaluating AI models through diverse data training projects. My work has encompassed both the creation of detailed rubric guidelines to refine AI behavior and the meticulous assessment of AI response quality. This experience has provided me with a deep understanding of the intricacies involved in data labeling and evaluation, particularly in ensuring AI models produce accurate, relevant, and contextually appropriate outputs. I've developed a keen eye for detail and a strong ability to translate complex requirements into actionable labeling strategies, contributing directly to the advancement of cutting-edge AI technologies.

Labeling Experience

Outlier

Scale AITextClassificationText Generation

This project aimed to refine and align a large language model (LLM) towards generating ideal, contextually relevant, and high-quality responses to complex user prompts. The primary focus was to develop robust evaluation frameworks and provide detailed feedback to steer the model's behavior. This involved creating comprehensive rubrics, generating diverse and challenging prompts, and rigorously assessing the model's output against defined quality standards. The project's overarching goal was to significantly improve the LLM's performance in understanding nuanced user requests and producing accurate, helpful, and ethical responses. My core responsibilities involved: * Complex Prompt Generation *Rubric Creation  * AI Response Rating and Classification The project involved a substantial volume of data.

This project aimed to refine and align a large language model (LLM) towards generating ideal, contextually relevant, and high-quality responses to complex user prompts. The primary focus was to develop robust evaluation frameworks and provide detailed feedback to steer the model's behavior. This involved creating comprehensive rubrics, generating diverse and challenging prompts, and rigorously assessing the model's output against defined quality standards. The project's overarching goal was to significantly improve the LLM's performance in understanding nuanced user requests and producing accurate, helpful, and ethical responses. My core responsibilities involved: * Complex Prompt Generation *Rubric Creation  * AI Response Rating and Classification The project involved a substantial volume of data.

2024

Education

Work History