For employers

Hire this AI Trainer

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

Invite to Job
Mahmoud Taha

Mahmoud Taha

LLM Evaluation and Text Generation Specialist & Versatile image & video.

Egypt flagCairo, Egypt
$20.00/hrExpertAws SagemakerAnno MageAppen

Key Skills

Software

AWS SageMakerAWS SageMaker
Anno-MageAnno-Mage
AppenAppen
ClickworkerClickworker
Data Annotation TechData Annotation Tech
EncordEncord
Img Lab
LabelboxLabelbox
Label StudioLabel Studio
Mighty AIMighty AI
OneFormaOneForma
RemotasksRemotasks
Scale AIScale AI
Surge AISurge AI
TolokaToloka
TelusTelus
V7 LabsV7 Labs
CVATCVAT

Top Subject Matter

No subject matter listed

Top Data Types

AudioAudio
TextText
VideoVideo

Top Task Types

Bounding Box
Classification
Data Collection
Evaluation Rating
Segmentation

Freelancer Overview

I have practical experience working on data labeling and AI training projects across images, videos, text, audio, and 3D data. My work has included video classification, image and video object annotation, evaluating AI responses, OCR text transcription, audio labeling for speech and noise, and 3D object annotation. These projects helped me build strong attention to detail and accuracy, while also giving me the ability to handle complex datasets. I enjoy contributing to AI model training pipelines by delivering high-quality labeled data that improves performance and reliability.

ExpertArabicEnglish

Labeling Experience

Data Annotation Tech

AI response evaluation

Data Annotation TechTextClassificationText Generation
Evaluated AI-generated responses for correctness, relevance, and clarity as part of Reinforcement Learning with Human Feedback (RLHF). Contributed to fine-tuning datasets by providing structured ratings and feedback to improve model alignment and usability.

Evaluated AI-generated responses for correctness, relevance, and clarity as part of Reinforcement Learning with Human Feedback (RLHF). Contributed to fine-tuning datasets by providing structured ratings and feedback to improve model alignment and usability.

2024 - 2024
Scale AI

LLM Math Problem write and solvin

Scale AITextClassificationRLHF
I worked on creating and reviewing math problem sets to support large language model training and evaluation. This included writing clear questions at different difficulty levels, checking step-by-step solutions, and making sure they followed standard math approaches. The project helped improve the model’s ability to understand and solve mathematical problems accurately.

I worked on creating and reviewing math problem sets to support large language model training and evaluation. This included writing clear questions at different difficulty levels, checking step-by-step solutions, and making sure they followed standard math approaches. The project helped improve the model’s ability to understand and solve mathematical problems accurately.

2023 - 2024
CVAT

Video Object Annotation

CVATVideoBounding BoxPolygon
Annotated 3D point cloud data with cuboids and segmentation for object detection tasks. Focused on labeling vehicles, pedestrians, and road elements with high spatial accuracy to support autonomous driving datasets.

Annotated 3D point cloud data with cuboids and segmentation for object detection tasks. Focused on labeling vehicles, pedestrians, and road elements with high spatial accuracy to support autonomous driving datasets.

2023 - 2023
Appen

OCR text transcription and validation

AppenImageBounding BoxObject Detection
Labeled and validated text extracted from scanned documents, correcting OCR errors and transcribing content accurately. Ensured entity-level tagging and classification to support high-quality document recognition models.

Labeled and validated text extracted from scanned documents, correcting OCR errors and transcribing content accurately. Ensured entity-level tagging and classification to support high-quality document recognition models.

2023 - 2023
Labelbox

Image object annotation

LabelboxImageBounding BoxPolygon
Performed object annotation in images using bounding boxes and polygons for detection and segmentation tasks. Ensured accuracy in labeling multiple objects per image and maintained high consistency across large datasets to support AI model performance.

Performed object annotation in images using bounding boxes and polygons for detection and segmentation tasks. Ensured accuracy in labeling multiple objects per image and maintained high consistency across large datasets to support AI model performance.

2022 - 2022

Education

A

Ain Shams University

Bachelor's Degree, Design And Production Mechanical Engineering

Bachelor's Degree
2019 - 2025

Work History

R

Regina for Pasta & Food Industries

Lean Six Sigma Green Belt Engineer

N/A
2025 - Present
O

Orascom Construction

Safety Training Participant

N/A
2025 - 2025