Shady Rafla

Key Skills

Software

CVAT

Labelbox

Label Studio

SuperAnnotate

V7 Labs

Top Subject Matter

No subject matter listed

Top Data Types

Image

Text

Video

Top Task Types

Bounding Box

Classification

Object Detection

Prompt Response Writing SFT

Tracking

Freelancer Overview

AI Training Expert with 2+ years in image and video annotation, delivering high-quality labeled data for machine learning and computer vision projects. Skilled in model training workflows and annotation tools. Experienced in most of AI training platforms such as (V7-CVAT-SuperAnnotate-Labelbox-Label Studio)

Labeling Experience

Human Activity & Identity Tagging from Police Security Footage

V7 LabsVideoBounding BoxTracking

Annotated surveillance videos from police security cams using V7 to identify and track individuals across multiple frames. Applied bounding boxes to persons of interest and tagged behaviors (e.g., walking, running, loitering). Ensured temporal consistency and accurate identity tracking in crowded urban environments.

Annotated surveillance videos from police security cams using V7 to identify and track individuals across multiple frames. Applied bounding boxes to persons of interest and tagged behaviors (e.g., walking, running, loitering). Ensured temporal consistency and accurate identity tracking in crowded urban environments.

2025 - 2025

Arabic Audio Transcription & Annotation for Gemini LLM Training

Label StudioAudioAudio Recording

Worked on transcribing and annotating native Arabic audio clips for Gemini's language model development. Tasks included accurate transcription, speaker identification, and tagging named entities (people, locations, organizations) across multiple dialects. Ensured high linguistic fidelity, alignment with LLM training protocols, and quality control with peer review to meet model training accuracy standards.

Worked on transcribing and annotating native Arabic audio clips for Gemini's language model development. Tasks included accurate transcription, speaker identification, and tagging named entities (people, locations, organizations) across multiple dialects. Ensured high linguistic fidelity, alignment with LLM training protocols, and quality control with peer review to meet model training accuracy standards.

2024 - 2024

Image & Text Toxicity Classification

LabelboxTextClassification

Reviewed and labeled sensitive content (text and images) to detect hate speech, nudity, and violence. Played a key role in developing safer content filtering systems for social platforms.

Reviewed and labeled sensitive content (text and images) to detect hate speech, nudity, and violence. Played a key role in developing safer content filtering systems for social platforms.

2024 - 2024

Crop Health Monitoring via Robot Imagery

CVATImageBounding Box

Annotated robot images of farmland to identify crop types, weed growth, and disease symptoms. Supported AI training for smart farming solutions to improve yield and resource efficiency.

Annotated robot images of farmland to identify crop types, weed growth, and disease symptoms. Supported AI training for smart farming solutions to improve yield and resource efficiency.

2023 - 2023

Vehicle & Pedestrian Detection for Self-Driving Cars

CVATImageBounding BoxPolygon

Labeled thousands of traffic scenes including cars, pedestrians, traffic lights, and lane markings for an autonomous driving system. Ensured frame consistency in video sequences and followed strict quality guidelines to meet safety-critical standards.

Labeled thousands of traffic scenes including cars, pedestrians, traffic lights, and lane markings for an autonomous driving system. Ensured frame consistency in video sequences and followed strict quality guidelines to meet safety-critical standards.

2022 - 2022

Education

Work History