Iulian Petru Ciubotărița

Key Skills

Software

Appen

Clickworker

Data Annotation Tech

Labelbox

OneForma

Remotasks

Scale AI

Top Subject Matter

No subject matter listed

Top Data Types

Audio

Text

Video

Top Task Types

Audio Recording

Prompt + Response Writing (SFT)

Text Generation

Freelancer Overview

I have hands-on experience in data labeling and AI training, with a strong focus on natural language processing and multilingual evaluation. My work on Project Babel involved LLM evaluation, where I assessed the performance, relevance, and coherence of large language model outputs across a range of prompts and tasks. I also contributed to Tansy – Translation QA, performing detailed quality assurance for English–Romanian translations, ensuring linguistic accuracy, fluency, and cultural relevance. Additionally, I participated in Stratus_Drake_Evals, focusing on structured evaluation tasks, and contributed to Resiliency Assessment, supporting the development of more robust AI systems. My bilingual proficiency in English and Romanian, along with a sharp eye for detail and consistency, allows me to deliver high-quality annotations and evaluations across diverse AI training contexts. I’m confident in my ability to support complex language tasks, cross-lingual projects, and model alignment efforts.

Labeling Experience

Resiliency Assessment

AppenTextClassificationPrompt Response Writing SFT

Resiliency Assessment focused on evaluating the robustness and reliability of AI model outputs, particularly in challenging or edge-case scenarios. My role involved identifying model vulnerabilities, assessing response stability across adversarial or ambiguous prompts, and ensuring consistency in alignment with safety and ethical guidelines. This work contributed to strengthening the model's ability to handle complex inputs while maintaining accuracy, coherence, and appropriate behavior under stress conditions.

Resiliency Assessment focused on evaluating the robustness and reliability of AI model outputs, particularly in challenging or edge-case scenarios. My role involved identifying model vulnerabilities, assessing response stability across adversarial or ambiguous prompts, and ensuring consistency in alignment with safety and ethical guidelines. This work contributed to strengthening the model's ability to handle complex inputs while maintaining accuracy, coherence, and appropriate behavior under stress conditions.

2024

Tansy - Translation QA

AppenTextTranslation LocalizationPrompt Response Writing SFT

Tansy – Translation QA focused on the quality assurance of machine-translated content, with an emphasis on English–Romanian language pairs. My responsibilities included reviewing and evaluating translations for accuracy, fluency, consistency, and cultural appropriateness. I provided detailed linguistic feedback and corrections to ensure high-quality output suitable for production environments. This project required strong bilingual skills, attention to detail, and familiarity with translation standards and localization best practices.

Tansy – Translation QA focused on the quality assurance of machine-translated content, with an emphasis on English–Romanian language pairs. My responsibilities included reviewing and evaluating translations for accuracy, fluency, consistency, and cultural appropriateness. I provided detailed linguistic feedback and corrections to ensure high-quality output suitable for production environments. This project required strong bilingual skills, attention to detail, and familiarity with translation standards and localization best practices.

2024

Babel - LLM Evaluation

AppenTextPrompt Response Writing SFT

Babel – LLM Evaluation involved evaluating the performance and outputs of large language models across a variety of prompts and tasks. My role centered on assessing response quality in terms of relevance, coherence, factual accuracy, and linguistic appropriateness. The project required applying detailed rubric-based scoring and written feedback to help improve model alignment and performance, especially in multilingual contexts including English and Romanian. This work supported the refinement of LLM behavior for more accurate, safe, and useful AI systems.

Babel – LLM Evaluation involved evaluating the performance and outputs of large language models across a variety of prompts and tasks. My role centered on assessing response quality in terms of relevance, coherence, factual accuracy, and linguistic appropriateness. The project required applying detailed rubric-based scoring and written feedback to help improve model alignment and performance, especially in multilingual contexts including English and Romanian. This work supported the refinement of LLM behavior for more accurate, safe, and useful AI systems.

2024

stratus_drake_evals

Scale AITextPrompt Response Writing SFT

Stratus_Drake_Evals focused on the structured evaluation of large language models (LLMs) across diverse linguistic and task-based scenarios. My role involved assessing model outputs for accuracy, coherence, and alignment with human expectations, particularly in nuanced reasoning and instruction-following tasks. I applied detailed evaluation guidelines to ensure consistency and quality, contributing to the development of benchmarks that inform model improvements. This project emphasized precision, critical thinking, and familiarity with AI evaluation frameworks.

Stratus_Drake_Evals focused on the structured evaluation of large language models (LLMs) across diverse linguistic and task-based scenarios. My role involved assessing model outputs for accuracy, coherence, and alignment with human expectations, particularly in nuanced reasoning and instruction-following tasks. I applied detailed evaluation guidelines to ensure consistency and quality, contributing to the development of benchmarks that inform model improvements. This project emphasized precision, critical thinking, and familiarity with AI evaluation frameworks.

2024

Education

Work History