Tarek Morchid

Key Skills

Software

Scale AI

Top Subject Matter

No subject matter listed

Top Data Types

Document

Image

Text

Top Task Types

Classification

Data Collection

Prompt Response Writing SFT

RLHF

Text Generation

Freelancer Overview

Expert in text analysis with geocomponents, with specialized experience across Belgium and Canada regions. I have extensive background in RLHF (Reinforcement Learning from Human Feedback) projects, developing strong skills in nuanced language evaluation and model improvement through human feedback. I also have dedicated experience with AI safety and content moderation projects, including the Pangolin initiative focused on identifying and evaluating dangerous, harmful, and heinous phrases generated by language models. This specialized safety work has honed my ability to make complex, contextually aware judgments on sensitive content while maintaining professional objectivity. My multilingual expertise (native French, C1 English, B2 Dutch, B1+ Spanish) enables me to provide culturally nuanced evaluation across linguistic contexts, making me particularly valuable for projects requiring cross-regional and cross-cultural assessment.

Labeling Experience

Pangolin project

Scale AITextEmotion RecognitionEvaluation Rating

AI safety work evaluating dangerous and harmful phrases in LLM outputs. Assessed model responses using a multi-dimensional evaluation framework with detailed feedback on high-risk language patterns to enhance AI safety and responsible model development.

AI safety work evaluating dangerous and harmful phrases in LLM outputs. Assessed model responses using a multi-dimensional evaluation framework with detailed feedback on high-risk language patterns to enhance AI safety and responsible model development.

2025 - 2025

RLHF project

Scale AITextClassificationText Generation

The LLM would be given a prompt about aspects or history of certain regions or countries such as Belgium and Canada, I would then need to factually verify quality on 7 evaluation criterias and rewrite if both response quality was bad.

The LLM would be given a prompt about aspects or history of certain regions or countries such as Belgium and Canada, I would then need to factually verify quality on 7 evaluation criterias and rewrite if both response quality was bad.

2024 - 2025

Education

Work History