For employers

Hire this AI Trainer

Sign in or create an account to invite AI Trainers to your job.

Invite to Job
Z
Zeed A.

Zeed A.

AI Research & Evaluation (Arabic Sign Language Recognition)

Turkey flagOnline, Turkey

Key Skills

Software

No software listed

Top Subject Matter

Arabic Sign Language Recognition (Computer Vision/Machine Learning)
Real-time Sign Language Recognition (Computer Vision)
Legal Services & Contract Review

Top Data Types

ImageImage
TextText
DocumentDocument

Top Task Types

No task types listed

Freelancer Overview

AI Research & Evaluation (Arabic Sign Language Recognition). Brings 3+ years of professional experience across legal operations, contract review, compliance, and structured analysis. Core strengths include Kaggle and Streamlit. Education includes Bachelor of Science, Istanbul Bahcesehir University (2022). AI-training focus includes data types such as Computer Code and Programming and labeling workflows including Evaluation, Rating, and Computer Programming.

Labeling Experience

Lead Developer & AI Researcher (ArSignTutor project)

Developed and built an AI application for webcam-based Arabic sign language recognition. Used deep learning and computer vision components to process visual input and run real-time inference. Delivered an interactive demo using an AI application framework. • Built a real-time Streamlit app for webcam sign recognition • Implemented model pipeline using MobileNetV2 and MediaPipe • Worked on dataset engineering to support the recognition system • Produced a working inference application for demonstration

2024 - 2024

AI Research & Evaluation (Arabic Sign Language Recognition)

Conducted AI evaluation and data quality checks as part of an Arabic sign language recognition project. Identified and investigated data leakage that inflated reported model accuracy. Implemented corrective methodology to improve the reliability of train-test separation. • Researched model evaluation approaches for sign language recognition • Detected severe data leakage inflating apparent accuracy • Designed dHash-based clustering to reduce train-test contamination • Verified improvements to reach honest test accuracy (94.33%)

2024 - 2024

Education

I

Istanbul Bahcesehir University

Bachelor of Science, Artificial Intelligence

Bachelor of Science
2022

Work History

A

Ali

Odoo ERP Administrator

Online
2026 - Present
R

Real Estate Office

Arabic–Turkish Translator

Istanbul
2024 - 2024