Prem S.

Key Skills

Software

Scale AI

Mercor

Top Subject Matter

Machine Learning Model Training Data

Employee Burnout Detection Data

Handwriting Replication Model Training Data

Top Data Types

Document

Image

Text

Top Task Types

Classification

Bounding Box

Segmentation

Object Detection

Transcription

Evaluation/Rating

Prompt + Response Writing (SFT)

Text Summarization

Question Answering

Text Generation

Freelancer Overview

AI/ML Intern — Data Annotation and Validation. Brings 1+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Internal, Proprietary Tooling, and Python. Education includes Master of Computer Applications, University of Petroleum & Energy Studies (2026) and Bachelor of Computer Applications, Maharishi Markandeshwar (Deemed) University (2023). AI-training focus includes data types such as Document and Image and labeling workflows including Classification.

Labeling Experience

AI/ML Intern — Data Annotation and Validation

DocumentClassification

As an AI/ML Intern at Xebia, I curated, cleaned, and validated structured datasets, focusing on annotation quality, label consistency, and data accuracy. I evaluated machine learning model outputs against ground truth data and provided structured feedback for model improvement. My work contributed to a measured improvement in model accuracy and maintained rigorous annotation standards. • Managed data curation and annotation for large, structured ML datasets. • Ensured label consistency and removed noise across datasets. • Performed quality validation and defect identification on annotated data. • Provided clear, actionable feedback to enhance ML model performance.

2026 - Present

ML Dataset Curation — Employee Burnout Detection

DocumentClassification

I designed and executed a comprehensive data annotation pipeline for an Employee Burnout Detection ML system. Through structured labeling and validation, I transformed raw behavioral data into refined supervised learning datasets. I used SHAP explainability to iteratively refine data quality and improve model predictions. • Developed an 8-step data annotation and data cleaning workflow. • Labeled, validated, and structured behavioral datasets for ML training. • Used SHAP explainability for label gap analysis and quality control. • Enhanced overall label quality for more robust model outputs.

2025 - 2025

Deep Learning Training Data — Handwriting Replication Model

ImageClassification

I curated and annotated a handwriting dataset, applying augmentation techniques to expand label variety for deep learning tasks. Annotation refinement at each training epoch ensured consistently realistic outputs for handwriting replication. My process directly improved the generalization and performance of deep learning models for text generation. • Managed dataset curation and annotation for custom handwriting samples. • Implemented data augmentation for improved diversity and model training. • Reviewed and refined annotations using iterative deep learning feedback. • Supported realistic, high-quality output generation for handwriting models.

2024 - 2024

Education

Work History