Mohamad Iqbal

Key Skills

Software

Other

Appen

Top Subject Matter

Video Scene-Cut & Action Annotation

Computer Vision/Object Detection

Text Classification & Intent Labeling

Top Data Types

Text

Image

Video

Audio

Document

Top Label Types

Action Recognition

Bounding Box

Classification

Evaluation Rating

Geocoding

Object Detection

Polygon

Tracking

RLHF

Transcription

Freelancer Overview

AI Training Contributor. Core strengths include Other and Appen. AI-training focus includes data types such as Video, Image, and Text and labeling workflows including Action Recognition, Object Detection, and Classification.

Labeling Experience

Data Annotator

AppenVideoSegmentation

I performed action labeling and scene annotation on video datasets at Appen for specialized ML training. This included identifying key activities and segmenting scenes. My contributions ensured a high standard of annotation accuracy and utility for machine learning purposes. • Annotated video actions and scenes. • Time-stamped activity boundaries. • Checked annotation consistency. • Followed task-specific labeling protocols.

I performed action labeling and scene annotation on video datasets at Appen for specialized ML training. This included identifying key activities and segmenting scenes. My contributions ensured a high standard of annotation accuracy and utility for machine learning purposes. • Annotated video actions and scenes. • Time-stamped activity boundaries. • Checked annotation consistency. • Followed task-specific labeling protocols.

2026 - Present

Visual Media Evaluation Rater

OtherVideoEvaluation RatingAction Recognition

Performed high-fidelity perceptual evaluations for generative AI and visual reconstruction models. The core tasks involved conducting meticulous side-by-side (SBS) comparisons between original media and AI-reconstructed videos or images. Evaluated and rated the closest match based on microscopic visual details, temporal consistency, and overall fidelity. These precise quality ratings provided critical human preference data required to fine-tune and improve the model's visual generation capabilities.

Performed high-fidelity perceptual evaluations for generative AI and visual reconstruction models. The core tasks involved conducting meticulous side-by-side (SBS) comparisons between original media and AI-reconstructed videos or images. Evaluated and rated the closest match based on microscopic visual details, temporal consistency, and overall fidelity. These precise quality ratings provided critical human preference data required to fine-tune and improve the model's visual generation capabilities.

2026 - Present

Visual Masking & Segmentation Validator

ImageSegmentation

Conducted rigorous quality assurance for image segmentation datasets. The primary task involved meticulously validating the accuracy and alignment of existing image masks. Furthermore, performed mask elimination tasks by analyzing specific contextual clues to identify and remove incorrect or irrelevant segmentations. Adhered to strict pixel-level quality guidelines to ensure the highest fidelity of the training data.

Conducted rigorous quality assurance for image segmentation datasets. The primary task involved meticulously validating the accuracy and alignment of existing image masks. Furthermore, performed mask elimination tasks by analyzing specific contextual clues to identify and remove incorrect or irrelevant segmentations. Adhered to strict pixel-level quality guidelines to ensure the highest fidelity of the training data.

2026 - Present

Text Classification & Intent Labeling Specialist

TextClassification

Conducted text classification and intent labeling tasks for diverse text datasets. The primary scope involved reading, analyzing, and accurately categorizing text data into predefined classes based on user intent. Adhered strictly to detailed project guidelines to ensure high accuracy in intent recognition, directly contributing to the training and refinement of Natural Language Processing (NLP) models.

Conducted text classification and intent labeling tasks for diverse text datasets. The primary scope involved reading, analyzing, and accurately categorizing text data into predefined classes based on user intent. Adhered strictly to detailed project guidelines to ensure high accuracy in intent recognition, directly contributing to the training and refinement of Natural Language Processing (NLP) models.

2026 - Present

AI Model RLHF Contributor

TextRLHF

Participated in a comprehensive Reinforcement Learning from Human Feedback (RLHF) project aimed at improving Large Language Model (LLM) performance. The core tasks involved meticulously evaluating and comparing AI-generated responses using structured rubrics. I conducted rigorous prompt ranking, factual consistency checks, and quality scoring. Strictly adhered to quality measures by reviewing all AI outputs for safety, relevance, and clarity to ensure the model aligns with human values and guidelines.

Participated in a comprehensive Reinforcement Learning from Human Feedback (RLHF) project aimed at improving Large Language Model (LLM) performance. The core tasks involved meticulously evaluating and comparing AI-generated responses using structured rubrics. I conducted rigorous prompt ranking, factual consistency checks, and quality scoring. Strictly adhered to quality measures by reviewing all AI outputs for safety, relevance, and clarity to ensure the model aligns with human values and guidelines.

2026 - Present

Show more

Education

Work History