Resume

University of Arizona

Master of Science in Data Science
Specialization in Machine Learning

Shiv Nadar University

Bachelor of Technology in Computer Science
Two times Dean's List Awardee
Specialization in Data Science and Big Data Analytics

HCLTech · Software Engineer

Generative AI Development Team

• Drove a 26% improvement in backend throughput (2,300 to 1,700 daily incidents) and a 76% cut in mean time to resolution (105 to 25 min), measured by ServiceNow KPI dashboards, by running scenario-based testing on ticket and content-quality time series and implementing Python, C++, and TypeScript validation pipelines for 4,000+ knowledge-base documents.
• Achieved a 35% reduction in document-quality escalations (post-release defect tickets) by performing what-if analysis on freshness, schema drift, and missing-field scenarios, and extending automated data validation with anomaly detection across federated enterprise sources.
• Delivered 100% on-time quarterly safety audit completion for finance and operations by building ServiceNow integrations, reproducible QA exports, and Excel-based control summaries with traceable assumptions.

HCLTech · Software Engineer Intern

• Drove adoption of sales performance tooling across HR, PMO, and L2 leadership by conducting EDA on stale metrics and workflow gaps, documenting metric logic, and shipping a React, TypeScript, and MS SQL Server application with leadership-ready views.
• Cut manual executive reporting effort by 50% by engineering reusable SQL features and views, then delivering Power BI and Tableau dashboards after exploratory reviews of KPI distributions, seasonality, and data-quality impacts.

Carrier · Web Developer Intern

• Achieved a 38% reduction in accounts-payable helpdesk volume (450 to 280 tickets/week), measured by ticketing analytics, by integrating an ML-driven FAQ chatbot with validation hooks and tuning retrieval quality with stakeholders across structured and messy input scenarios.
• Drove a 35% increase in in-app engagement by designing React, JavaScript, and jQuery workflows with Excel-backed operational reporting and code reviews that kept data contracts stable for analytics consumers.

PricewaterhouseCoopers · Technical Intern

• Enabled sub-second retrieval for 1,000+ employee records (Neo4j benchmarks) by running exploratory cleaning on CNCF exports, engineering graph features in Python, and validating model assumptions through comparison testing before scale-up.
• Accelerated the month-end reporting cycle by 40% by building Python and REST data pipelines, automating validation, and pairing exploratory variance and outlier analysis in Excel with concise summaries for non-technical partners.

FaceForge Attendance System

Python, HTML, CSS, React, Tailwind CSS, PostgreSQL

• Developed the attendance system using Python and OpenCV for educational and corporate settings.
• Used dlib to create facial templates for recognition and integrated a proxy detector using object detection.
• Developed the UI using React and Tailwind CSS; maintained records on PostgreSQL.

Self-Correction for Human Parsing Model

Python, Git

• Compared the SCHP model with other single-person human parsing state-of-the-art methods.
• Evaluated on benchmarks: Look into Person (LIP), Active Template Regression, and Pascal Part.
• Achieved 84.7% mIoU on LIP dataset, 2.3% higher than the original SCHP paper result.

ChatGPT Plagiarism Detector

Python, EasyOCR, APIs

• Developed plagiarism detection tool using ChatGPT API & paraphrasing API.
• Implemented cosine similarity with vector space model to detect paraphrased text.
• Performed OCR with EasyOCR; tested on 30 docs with 93% accuracy.

International Conference on Database and Expert Systems Applications (DEXA) 2024, Naples, Italy