Ahmed Marjawi

Key Skills

Software

Appen

Data Annotation Tech

Remotasks

Scale AI

Top Subject Matter

No subject matter listed

Top Data Types

Audio

Computer Code Programming

Text

Top Label Types

Audio Recording

Computer Programming Coding

Data Collection

Text Generation

Text Summarization

Freelancer Overview

I have hands-on experience in data labeling and AI training data, specializing in coding, audio, and text datasets. My technical background in programming enables me to efficiently handle large volumes of complex data, ensuring accurate annotation and preprocessing to improve AI model performance. I am skilled in working with diverse data formats, including audio transcriptions and text classification, which allows me to contribute effectively to natural language processing (NLP) and speech recognition projects. My attention to detail and understanding of coding principles set me apart in identifying subtle data nuances and maintaining high-quality standards throughout the labeling process. I am comfortable using various annotation tools and following precise guidelines to deliver reliable training datasets that enhance the accuracy and robustness of machine learning models.

Labeling Experience

Code Snippet Classification and Function Annotation for Automated Code Review

Scale AIComputer Code ProgrammingClassificationComputer Programming Coding

Annotated and classified over 5,000 code snippets across multiple programming languages (Python, JavaScript, C++) by function type, complexity, and potential code smells. Created detailed labels for function calls and dependencies to improve machine learning models focused on automated code review and bug detection. Also contributed to prompt and response writing for supervised fine-tuning (SFT) of coding assistants. Maintained annotation accuracy through peer review and continuous guideline updates.

Annotated and classified over 5,000 code snippets across multiple programming languages (Python, JavaScript, C++) by function type, complexity, and potential code smells. Created detailed labels for function calls and dependencies to improve machine learning models focused on automated code review and bug detection. Also contributed to prompt and response writing for supervised fine-tuning (SFT) of coding assistants. Maintained annotation accuracy through peer review and continuous guideline updates.

2024 - 2024

Audio Transcription and Labeling for Voice Assistant Development

Scale AIAudioEntity Ner ClassificationClassification

Performed detailed transcription and entity tagging on a dataset of over 10,000 voice commands collected from diverse speakers. Annotated audio clips with speaker identification, background noise classification, and command intent labeling to train voice assistant NLP models. Quality assurance processes included audio quality checks, timestamp accuracy validation, and adherence to annotation guidelines. This dataset helped enhance the model’s ability to understand natural speech variations and accents.

Performed detailed transcription and entity tagging on a dataset of over 10,000 voice commands collected from diverse speakers. Annotated audio clips with speaker identification, background noise classification, and command intent labeling to train voice assistant NLP models. Quality assurance processes included audio quality checks, timestamp accuracy validation, and adherence to annotation guidelines. This dataset helped enhance the model’s ability to understand natural speech variations and accents.

2024 - 2024

Text Annotation for Sentiment Analysis of Customer Feedback

Scale AIImageClassificationEmotion Recognition

Led the annotation of over 50,000 customer reviews for sentiment classification and emotion recognition, categorizing texts into positive, negative, neutral, and further emotional states like frustration, satisfaction, or confusion. Tasks included highlighting key phrases for summarization and generating concise summaries of long feedback to improve model training for automated sentiment detection. Ensured high data quality by conducting double-blind annotations and reconciliation of discrepancies, achieving a 98% inter-annotator agreement.

Led the annotation of over 50,000 customer reviews for sentiment classification and emotion recognition, categorizing texts into positive, negative, neutral, and further emotional states like frustration, satisfaction, or confusion. Tasks included highlighting key phrases for summarization and generating concise summaries of long feedback to improve model training for automated sentiment detection. Ensured high data quality by conducting double-blind annotations and reconciliation of discrepancies, achieving a 98% inter-annotator agreement.

2024 - 2024

Education

Work History