Angelina

Key Skills

Software

Scale AI

Top Subject Matter

No subject matter listed

Top Data Types

Text

Top Task Types

Translation Localization

RLHF

Evaluation Rating

Prompt Response Writing SFT

Freelancer Overview

I am a multilingual professional with a strong background in philology, media and communication, specializing in AI training data, data annotation, and evaluation of large language models (LLMs). My experience includes contributing to post-training projects such as Reinforcement Learning from Human Feedback (RLHF) and Supervised Fine-Tuning (SFT), where I analyzed, evaluated, and improved model-generated responses for accuracy, clarity, and safety. I am skilled in text annotation, content review, and localization, with a proven ability to assess model behavior on sensitive content in alignment with safety and ethics guidelines. My attention to detail and adaptability enable me to deliver high-quality results on large-scale text datasets in fast-paced, deadline-driven environments. I am proficient in English, native Russian speaker, and also speak Korean, and French, and I excel at ensuring linguistic accuracy and consistency across multilingual data projects.

Labeling Experience

Russian Language Specialist — LLM Training & Evaluation

Scale AITextTranslation LocalizationRLHF

Worked as a Russian Language Specialist on ongoing large language model (LLM) training projects focused on improving model quality, safety, and linguistic accuracy. Responsibilities included text annotation, evaluation, and writing tasks across multiple project types, such as translation/localization, prompt–response writing, and model output assessment. Evaluated AI-generated responses against detailed guidelines and rubrics, focusing on instruction adherence, factual accuracy, clarity, tone, safety, and localization for Russian-speaking audiences.

Worked as a Russian Language Specialist on ongoing large language model (LLM) training projects focused on improving model quality, safety, and linguistic accuracy. Responsibilities included text annotation, evaluation, and writing tasks across multiple project types, such as translation/localization, prompt–response writing, and model output assessment. Evaluated AI-generated responses against detailed guidelines and rubrics, focusing on instruction adherence, factual accuracy, clarity, tone, safety, and localization for Russian-speaking audiences.

2024 - 2025

Education

Work History