Shoaib Wani

Key Skills

Software

Other

iMerit

Scale AI

CVAT

Top Subject Matter

No subject matter listed

Top Data Types

Text

Video

Image

Top Label Types

Evaluation Rating

Red Teaming

Computer Programming Coding

Prompt Response Writing SFT

Bounding Box

Classification

Transcription

Freelancer Overview

I am an AI and machine learning professional with 7 years of hands-on experience in data annotation, labeling, and high-quality AI training data creation across diverse domains including computer vision, NLP, medical imaging, and conversational AI. My work spans LLM response evaluation, RLHF feedback, gold label authoring, bias and hallucination detection, and rigorous dataset QA for leading organizations such as Outlier AI, Scale AI, and iMerit. I am highly skilled in Python, PyTorch, TensorFlow, and annotation tools like CVAT, with a strong focus on delivering accurate, reliable datasets for supervised learning and model alignment. My projects include developing multimodal annotation pipelines, medical image labeling for pneumonia detection, and designing evaluation frameworks for LLM safety and alignment, all grounded in a research-driven, human-centered approach. Product Listing Content and Visual Structuring for Web Interfaces. Selected UI and dataset visualization examples available on GitHub.

Labeling Experience

Computer Vision Annotation Contractor – CVAT Projects

CVATImageBounding Box

As a Computer Vision Annotation Contractor for CVAT Projects, I annotated bounding boxes and segmentation masks for object detection datasets. My work focused on labeling images to assist in the training of computer vision AI. This experience contributed to dataset quality and model precision. • Drew bounding boxes around objects in image datasets • Applied segmentation masks for detailed labeling • Reviewed and validated annotations for quality control • Used CVAT tools to ensure efficient annotation workflows •Reviewed visual data quality and object placement within images used for machine learning datasets.

As a Computer Vision Annotation Contractor for CVAT Projects, I annotated bounding boxes and segmentation masks for object detection datasets. My work focused on labeling images to assist in the training of computer vision AI. This experience contributed to dataset quality and model precision. • Drew bounding boxes around objects in image datasets • Applied segmentation masks for detailed labeling • Reviewed and validated annotations for quality control • Used CVAT tools to ensure efficient annotation workflows •Reviewed visual data quality and object placement within images used for machine learning datasets.

2023 - 2025

AI Data Annotation Specialist – iMerit

ImeritTextBounding BoxClassification

As an AI Data Annotation Specialist at iMerit, I performed high-accuracy annotation tasks on textual and image data for supervised machine learning projects. I maintained rigorous QA standards and labeled complex edge cases in datasets. My responsibility was to ensure data quality and consistency for effective model training. • Provided precise data labeling for both text and images • Upheld QA standards throughout annotation workflows • Identified and labeled ambiguous or challenging edge cases • Enhanced the accuracy of training data for ML systems

As an AI Data Annotation Specialist at iMerit, I performed high-accuracy annotation tasks on textual and image data for supervised machine learning projects. I maintained rigorous QA standards and labeled complex edge cases in datasets. My responsibility was to ensure data quality and consistency for effective model training. • Provided precise data labeling for both text and images • Upheld QA standards throughout annotation workflows • Identified and labeled ambiguous or challenging edge cases • Enhanced the accuracy of training data for ML systems

2023

LLM Evaluator – Outlier AI

OtherTextEvaluation RatingRed Teaming

As an LLM Evaluator at Outlier AI, I evaluated and ranked LLM-generated responses according to defined metrics. I authored high-quality gold standard responses and flagged instances of hallucination and bias in outputs. My work supported RLHF pipelines for improving natural language processing models. • Conducted detailed prompt response grading for accuracy and relevance • Developed clear, gold-standard responses for benchmarking • Identified and reported cases of bias and hallucination in model outputs • Collaborated with RLHF teams for model alignment improvements

As an LLM Evaluator at Outlier AI, I evaluated and ranked LLM-generated responses according to defined metrics. I authored high-quality gold standard responses and flagged instances of hallucination and bias in outputs. My work supported RLHF pipelines for improving natural language processing models. • Conducted detailed prompt response grading for accuracy and relevance • Developed clear, gold-standard responses for benchmarking • Identified and reported cases of bias and hallucination in model outputs • Collaborated with RLHF teams for model alignment improvements

2023

LLM Evaluation & Dataset QA Analyst – Scale AI

Scale AITextEvaluation Rating

As an LLM Evaluation & Dataset QA Analyst at Scale AI, I was responsible for grading LLM responses and validating gold-label annotations. I wrote and structured JSON prompts and performed preference ranking for conversational models. This work contributed to dataset quality assurance and model improvement. • Graded LLM outputs to assess correctness and coherence • Authored and reviewed JSON prompt-response pairs • Validated gold label consistency across datasets • Ranked and annotated preferences for dialogue systems

As an LLM Evaluation & Dataset QA Analyst at Scale AI, I was responsible for grading LLM responses and validating gold-label annotations. I wrote and structured JSON prompts and performed preference ranking for conversational models. This work contributed to dataset quality assurance and model improvement. • Graded LLM outputs to assess correctness and coherence • Authored and reviewed JSON prompt-response pairs • Validated gold label consistency across datasets • Ranked and annotated preferences for dialogue systems

2023

AI Training Data Contributor – Atlas AI

OtherVideoTranscription

At Atlas AI, I contributed to AI training datasets through video transcription and annotation using bounding boxes. This role involved converting spoken language into text and marking specific objects within video frames. The outputs were prepared for use in improving computer vision and speech recognition models. • Transcribed audio tracks from various video sources • Created bounding box annotations around relevant objects in videos • Ensured accuracy and consistency across data samples • Supported the development of robust AI applications

At Atlas AI, I contributed to AI training datasets through video transcription and annotation using bounding boxes. This role involved converting spoken language into text and marking specific objects within video frames. The outputs were prepared for use in improving computer vision and speech recognition models. • Transcribed audio tracks from various video sources • Created bounding box annotations around relevant objects in videos • Ensured accuracy and consistency across data samples • Supported the development of robust AI applications

2019 - 2021

Education

Work History