Mohamed Wael Oueslati

Key Skills

Software

CVAT

Labelbox

LabelImg

Prodigy

Label Studio

Top Subject Matter

No subject matter listed

Top Data Types

Image

Text

Video

Top Task Types

Bounding Box

Classification

Polygon

Text Generation

Translation Localization

Freelancer Overview

I am a Data Annotation Specialist and third-year Software Engineering student with a strong foundation in data preprocessing, text classification, and image annotation. Through self-driven projects like manually labeling 500+ tweets for sentiment analysis, completing CVAT polygon-segmentation and bounding-box exercises, and refining LLM prompts on Outlier.ai. I’ve built practical skills in quality assurance and tool workflows (CVAT, LabelStudio, LabelImg). Fluent in Arabic, French, and English (plus intermediate Turkish), I bring meticulous attention to detail and clear communication to every task. Eager to ramp up quickly in a professional AI-training environment, I’m passionate about delivering consistent, high-quality annotations and collaborating with cross-functional teams to support cutting-edge model development.

Labeling Experience

CVAT Polygon & Bounding-Box Tutorial

CVATImageBounding BoxPolygon

Completed CVAT’s official exercises, drawing bounding boxes around objects and creating precise polygon masks on sample images. Exported annotations in COCO format and performed quality checks against provided ground-truth masks. Documented key workflows for future reference.

Completed CVAT’s official exercises, drawing bounding boxes around objects and creating precise polygon masks on sample images. Exported annotations in COCO format and performed quality checks against provided ground-truth masks. Documented key workflows for future reference.

2025 - 2025

Healthcare Data Preprocessing & Annotation

Label StudioTextEntity Ner ClassificationClassification

Cleaned a patient records dataset by resolving missing values and normalizing free-text disease descriptions into standardized categories. Grouped rare disease classes to improve sample balance, then validated label accuracy against source records. Prepared the cleaned dataset for downstream Random Forest and Naïve Bayes experiments.

Cleaned a patient records dataset by resolving missing values and normalizing free-text disease descriptions into standardized categories. Grouped rare disease classes to improve sample balance, then validated label accuracy against source records. Prepared the cleaned dataset for downstream Random Forest and Naïve Bayes experiments.

2025 - 2025

Social Media Sentiment Analysis Dataset

Label StudioTextClassificationEmotion Recognition

Collected 500+ real-world tweets via the Twitter API and manually labeled each as positive, neutral, or negative. Cleaned and preprocessed text by removing URLs, mentions, and emojis. Wrote detailed annotation guidelines and ran inter-annotator checks to ensure consistency. Exported a final CSV schema ready for model training.

Collected 500+ real-world tweets via the Twitter API and manually labeled each as positive, neutral, or negative. Cleaned and preprocessed text by removing URLs, mentions, and emojis. Wrote detailed annotation guidelines and ran inter-annotator checks to ensure consistency. Exported a final CSV schema ready for model training.

2025 - 2025

Education

Work History