Pfizer logoPfizer Externship via Extern
Yash Lalwani portrait

CS junior at Rutgers building AI projects for real documents

I create AI prototypes that pair OCR, LLMs, and RAG to extract and summarize information from enterprise PDFs, sharpening my resume for internships.

Work samples

Portfolio items include a Pfizer externship prototype that automates enterprise PDF workflows using OCR, LLMs, and RAG, plus related AI experiments aimed at resume-ready outcomes.

  • Pfizer Advanced: AI-Powered Document Insights & Data Extraction Externship
    Pfizer Advanced: AI-Powered Document Insights & Data Extraction Externship
    Pfizer logo

    Pfizer · ✅ Verified by Extern · ⏱️ In progress

    Pfizer Advanced: AI-Powered Document Insights & Data Extraction Externship

    Prototype AI-powered document intelligence with Pfizer—using OCR, LLMs, and RAG to automate real enterprise PDF workflows and build a standout portfolio project.

    AI & MLPythonDocument IntelligencePresentation Skills
  • MyGPT
    MyGPT

    Built a full-stack, AI-powered chat application integrating OpenAI’s GPT API for conversational capabilities. Implemented persistent chat history to maintain conversational context across multiple user sessions. Developed a responsive, user-friendly UI that emulates the ChatGPT interface for enhanced user experience.

    OpenAI APIJavaScriptMongoDBExpressReactNode.js

About me

I create AI prototypes that pair OCR, LLMs, and RAG to extract and summarize information from enterprise PDFs, sharpening my resume for internships.

I am a junior Computer Science student at Rutgers University focused on Artificial Intelligence. I am building hands-on experience through projects and a Pfizer externship that apply OCR, LLMs, and retrieval-augmented generation to real document workflows as I prepare for internships and early-career roles.

Externships

Pfizer Advanced: AI-Powered Document Insights & Data Extraction Externship

In progress

Experience

AI/ML Research Assistant & Automation Engineer

Rutgers University - Bloustein School · Oct 2025 – Present

AI Extern

Pfizer · Sep 2026 – Present

Model Validation Expert (MOVE) Fellow

Handshake AI · Nov 2025 – Dec 2025

Education

Rutgers University

Bachelor of Science in Computer Science, Minor in Data Science (Senior) · Class of 2027

Skills

PythonOCR pipelinesLarge language models (LLMs)Retrieval-augmented generation (RAG)Document data extractionMachine learning fundamentalsGit and version control

✅ Verified by Extern · ⏱️ In progress

Pfizer Advanced: AI-Powered Document Insights & Data Extraction Externship

Prototype AI-powered document intelligence with Pfizer—using OCR, LLMs, and RAG to automate real enterprise PDF workflows and build a standout portfolio project.

AI & MLPythonDocument IntelligencePresentation Skills

Overview

The work prototyped AI document-intelligence pipelines combining OCR, LLMs, and retrieval-augmented generation to extract structured data and surface insights from enterprise PDFs. Deliverables included prototypes and technical notes on model choices, architectures, and integration patterns. The work examined pharmaceutical vendor documents for extraction challenges and documented problematic

Pfizer Advanced: AI-Powered Document Insights & Data Extraction Externship

What I've accomplished

I analyzed vendor pharmaceutical documents and produced a detailed problem log that identified mixed date formats, dated records, and visual elements (tables, graphs, chemical diagrams, checkboxes) likely to impede OCR and automated extraction.

Project breakdown

I examined vendor documents, identified mixed date formats and dated records, and flagged visual elements (tables, graphs, chemical diagrams, checkboxes) that could hinder OCR and NLP. I noted potential confusion points for automated extraction.

View all work

MyGPT

Built a full-stack, AI-powered chat application integrating OpenAI’s GPT API for conversational capabilities. Implemented persistent chat history to maintain conversational context across multiple user sessions. Developed a responsive, user-friendly UI that emulates the ChatGPT interface for enhanced user experience.

OpenAI APIJavaScriptMongoDBExpressReactNode.js
View all work

Newsly.AI

Developed an AI-driven news platform using Python and Streamlit to enhance news exploration. Implemented sentiment analysis and visualizations (charts and word clouds) to extract insights from articles. Integrated an assistant chatbot to offer personalized news discovery and facilitate user interaction.

PythonStreamlitpandasscikit-learnMatplotlibWeb Scraping
View all work

Zerodha Clone

Created a responsive trading interface that replicates core functionality and workflows of the Zerodha platform. Implemented secure authentication to enable dashboard access and safeguard user data and transactions. Integrated Chart.js for real-time visualization of stock holdings and enabled buy/sell order execution workflows.

AuthNAuthZChart.jsReactNode.jsExpressMongoDB
View all work