I analyze complex healthcare and operational datasets to create actionable dashboards and risk indicators. My work empowers informed decisions across teams.
Dive into my portfolio showcasing advanced analytics projects, including predictive modeling and fraud detection, powered by SQL, Python, and Tableau.
.jpg)
Pfizer · Oct 2026 · ✅ Verified by Extern
Prototype AI-powered document intelligence with Pfizer—using OCR, LLMs, and RAG to automate real enterprise PDF workflows and build a standout portfolio project.
Built a time-aware XGBoost sales-prediction model on 420K+ weekly records (20+ engineered features), reaching 0.88 R² validated with TimeSeries Split, to support inventory and demand-planning decisions. Developed an XGBoost credit card fraud detection model on 1.85M transactions (0.5% fraud rate) us
I analyze complex healthcare and operational datasets to create actionable dashboards and risk indicators. My work empowers informed decisions across teams.
Data analyst with an engineering background who turns large healthcare, pharma, and financial datasets into clear, data- driven decisions using Python, SQL, Excel, Tableau, and Power BI. Built XGBoost models on up to 1.85M records and a fully local RAG system that raised answer accuracy to 88% and cut latency 71%. Led 2 teams to 1st place in hackathons.
Externships
Pfizer Advanced: AI-Powered Document Insights & Data Extraction Externship
Completed · Oct 2026
Experience
Data Science Extern
Data Science Extern · July 2026 to Present
Data Analyst, NumpyNinja
NumpyNinja · July 2025 to Present
Career Development and U.S. Relocation
Career Development and U.S. Relocation · August 2022 to May 2025
Engineer, Engineering and Application Support
Sterling E-Mobility · January 2022 to July 2022
Trainee, System Design
Philips India · September 2020 to July 2021
Education
National Institute of Technology, Kurukshetra, India
Master of Technology in Power System · Class of 2021
Government Rewa Engineering College (RGPV), India
Bachelor of Engineering in Electrical Engineering · Class of 2018
Skills
✅ Verified by Extern
Prototype AI-powered document intelligence with Pfizer—using OCR, LLMs, and RAG to automate real enterprise PDF workflows and build a standout portfolio project.
The work prototyped an AI-powered document intelligence pipeline combining OCR, LLMs, and retrieval-augmented generation to process enterprise PDFs. The project investigated LLM architectures and the components needed to extract, index, and query text from scanned documents. Deliverables included research notes and a technical explanation of how LLMs operate within the pipeline.
I produced a written summary that explained how LLMs learn from large text corpora, how they tokenize input and predict next tokens, how training and parameter updates work, and how fine-tuning with human feedback modifies model behavior.
Built a time-aware XGBoost sales-prediction model on 420K+ weekly records (20+ engineered features), reaching 0.88 R² validated with TimeSeries Split, to support inventory and demand-planning decisions. Developed an XGBoost credit card fraud detection model on 1.85M transactions (0.5% fraud rate) us
