Mohammed Elabridi

Key Skills

Software

CVAT

LabelImg

Label Studio

OpenCV AI Kit (OAK)

Remotasks

Scale AI

Top Subject Matter

No subject matter listed

Top Data Types

Audio

Computer Code Programming

Text

Top Task Types

Audio Recording

Data Collection

Evaluation Rating

Prompt Response Writing SFT

Text Generation

Freelancer Overview

With a strong academic background in engineering and AI, and handson experience developing multilingual chatbots and LLM evaluation pipelines, I’ve worked on a variety of NLP-related data labeling and training tasks. I specialize in text-based AI training, including prompt and response writing, dialogue evaluation, classification, and summarization, especially in Arabic, French, and English. I’ve contributed to multiple projects involving chatbot dialogue structuring, fine tuning LLMs, and evaluating AI responses for alignment and relevance. My experience includes manual annotation, entity classification, and prompt evaluation for low-resource languages. I’m familiar with tools such as Label Studio, LabelImg, and internal labeling pipelines used in academic and startup settings. My unique value lies in combining deep language awareness with technical know-how, especially for projects requiring regional language expertise (Arabic dialects) and human-centric AI evaluation.

Labeling Experience

Audio Transcription & Classification for Multilingual Voice Assistant

Label StudioAudioClassificationEmotion Recognition

Worked on a voice assistant training project that required transcribing and annotating audio recordings in Darija, French, and English. Tasks included manual transcription of user commands, labeling emotional tone, and classifying intent categories for each clip. In some cases, provided translations or cleaned noisy recordings for preprocessing. Ensured high accuracy and linguistic consistency across dialects. Also helped refine guidelines for labeling regional Arabic variations and participated in reviewing samples for quality assurance.

Worked on a voice assistant training project that required transcribing and annotating audio recordings in Darija, French, and English. Tasks included manual transcription of user commands, labeling emotional tone, and classifying intent categories for each clip. In some cases, provided translations or cleaned noisy recordings for preprocessing. Ensured high accuracy and linguistic consistency across dialects. Also helped refine guidelines for labeling regional Arabic variations and participated in reviewing samples for quality assurance.

2024 - 2025

Image & Video Annotation for Object Detection and Action Recognition

CVATVideoBounding BoxClassification

Labeled image and video datasets to support the training of computer vision models. Tasks involved drawing precise bounding boxes, classifying objects, and annotating sequences for action recognition. Also worked on multi-frame tracking across video clips to assist with behavior prediction and motion analysis. Ensured annotation consistency and followed project-specific labeling protocols. Collaborated with quality assurance teams and participated in iterative feedback cycles to improve annotation standards.

Labeled image and video datasets to support the training of computer vision models. Tasks involved drawing precise bounding boxes, classifying objects, and annotating sequences for action recognition. Also worked on multi-frame tracking across video clips to assist with behavior prediction and motion analysis. Ensured annotation consistency and followed project-specific labeling protocols. Collaborated with quality assurance teams and participated in iterative feedback cycles to improve annotation standards.

2023 - 2024

Multilingual Chatbot Evaluation & Prompt Annotation (Darija/French/English)

Label StudioTextEntity Ner ClassificationText Summarization

Worked on a project focused on building and evaluating multilingual chatbots for educational and customer support use cases. Tasks included writing high-quality prompts, rating LLM outputs, correcting chatbot dialogues, and annotating user intents and named entities in Darija, French, and English. Contributed to the fine-tuning and human feedback stages (RLHF). Helped establish quality guidelines for low-resource languages and worked closely with dev teams to optimize labeling efficiency.

Worked on a project focused on building and evaluating multilingual chatbots for educational and customer support use cases. Tasks included writing high-quality prompts, rating LLM outputs, correcting chatbot dialogues, and annotating user intents and named entities in Darija, French, and English. Contributed to the fine-tuning and human feedback stages (RLHF). Helped establish quality guidelines for low-resource languages and worked closely with dev teams to optimize labeling efficiency.

2023 - 2024

Education

Work History