Angela Meili

Key Skills

Software

SuperAnnotate

Telus

Top Subject Matter

No subject matter listed

Top Data Types

Text

Top Task Types

Classification

Emotion Recognition

Evaluation Rating

Prompt Response Writing SFT

Red Teaming

RLHF

Text Generation

Text Summarization

Translation Localization

Freelancer Overview

I am an experienced cognitive linguist and AI safety researcher with a strong background in data annotation, RLHF, and semantic red teaming for large language models. My work bridges theoretical expertise in pragmatics, frame semantics, and conceptual metaphor analysis with hands-on experience across leading AI training platforms such as Active Fence, DataAnnotation.tech, Remotasks, Telus International, and Invisible Technologies. I specialize in crafting and evaluating adversarial prompts, detecting pragmatic and cultural vulnerabilities in multilingual data (especially Portuguese and English), and ensuring AI outputs align with safety and fairness guidelines. My interdisciplinary approach combines advanced qualitative research methods with real-world annotation projects, making me adept at identifying nuanced issues in NLP data and contributing to responsible, bias-aware AI development.

Labeling Experience

RLHF Safety & Security Specialist / AI Content Evaluator

SuperannotateTextRLHFEvaluation Rating

Atuação no treinamento e refinamento de Modelos de Linguagem de Larga Escala (LLMs) através de Reinforcement Learning from Human Feedback (RLHF). O trabalho foca em Red Teaming para identificação de vulnerabilidades (jailbreaks e injeção de prompts), além da avaliação de respostas quanto à segurança, precisão factual e alinhamento ético. Aplico critérios rigorosos de qualidade e segurança (Safety & Security Behaviors) para garantir que o modelo opere dentro das diretrizes estabelecidas

Atuação no treinamento e refinamento de Modelos de Linguagem de Larga Escala (LLMs) através de Reinforcement Learning from Human Feedback (RLHF). O trabalho foca em Red Teaming para identificação de vulnerabilidades (jailbreaks e injeção de prompts), além da avaliação de respostas quanto à segurança, precisão factual e alinhamento ético. Aplico critérios rigorosos de qualidade e segurança (Safety & Security Behaviors) para garantir que o modelo opere dentro das diretrizes estabelecidas

2025

AI Community Search Evaluator / Data Analyst (Telus International)

TelusTextClassificationRLHF

Acted as an AI Community Search Evaluator via the Sigma platform for Telus International. The core responsibilities involved high-precision data labeling and classification to improve search engine algorithms and machine learning models. I performed deep analysis of search relevance, geospatial data verification, and semantic categorization, ensuring that the AI-generated outputs met strict quality, accuracy, and user-intent guidelines. This role required consistent adherence to complex project documentation and high performance metrics.

Acted as an AI Community Search Evaluator via the Sigma platform for Telus International. The core responsibilities involved high-precision data labeling and classification to improve search engine algorithms and machine learning models. I performed deep analysis of search relevance, geospatial data verification, and semantic categorization, ensuring that the AI-generated outputs met strict quality, accuracy, and user-intent guidelines. This role required consistent adherence to complex project documentation and high performance metrics.

2025 - 2025

Education

Work History