Omar Alazoni

Key Skills

Software

AWS SageMaker

CVAT

Label Studio

Roboflow

Top Subject Matter

No subject matter listed

Top Data Types

Computer Code Programming

Image

Video

Top Task Types

Bounding Box

Classification

Object Detection

Prompt Response Writing SFT

RLHF

Freelancer Overview

Robotics and AI engineer with experience in computer vision using YOLO and various algorithms for object detection (PPE, Fire, Person) and annotate the data using label-studio or make-sence tools. Proficient in training models, building and evaluating LLMs, prompt engineering, and developing AI agents and chatbots with LangGraph.

Labeling Experience

Fire detection

CVATVideoBounding BoxClassification

I collect and annotate data from Roboflow and my own sources (CCTV) to detect fire and flames. This system will trigger an alarm and notification upon detecting a fire or flame-related violation within the facility, enhancing safety.

I collect and annotate data from Roboflow and my own sources (CCTV) to detect fire and flames. This system will trigger an alarm and notification upon detecting a fire or flame-related violation within the facility, enhancing safety.

2025

PPE

Label StudioImageBounding BoxSegmentation

My responsibilities have included the collection and annotation of data form cctv camera for training modele to perform PPE compliance, and recognizing faces accurately. Project workflow: Collect data from real CCTV footage within the test location. Organize data and extract frames from test record videos. Annotate the data with Label Studio. Incorporate a dataset from Roboflow and reformat it to align with the class order of my custom data. Begin initial data training in a local environment and monitor the metrics. Train the model with the full dataset for 150 epochs using YOLO11 on Colab Pro (A100). Validate and test the model to ensure its efficiency. Test the model on real-time video recording. Future work: Deploy the model on a webpage for local or network use, creating a market solution.

My responsibilities have included the collection and annotation of data form cctv camera for training modele to perform PPE compliance, and recognizing faces accurately. Project workflow: Collect data from real CCTV footage within the test location. Organize data and extract frames from test record videos. Annotate the data with Label Studio. Incorporate a dataset from Roboflow and reformat it to align with the class order of my custom data. Begin initial data training in a local environment and monitor the metrics. Train the model with the full dataset for 150 epochs using YOLO11 on Colab Pro (A100). Validate and test the model to ensure its efficiency. Test the model on real-time video recording. Future work: Deploy the model on a webpage for local or network use, creating a market solution.

2025

Train Models (Face identify, PPE classification and detect, fire)

Opencv AI Kit OakImageBounding BoxSegmentation

Train a model for object detection and classification in diverse computer vision applications.

Train a model for object detection and classification in diverse computer vision applications.

2024

Education

Work History