Hello, I'm Anuj Shah

Computer Science student at the University of Texas at Austin. Experienced in Backend development, Operating systems, ML, and Cloud Computing. Passionate about building impactful software solutions. Work experience at JPMorgan Chase, UH Hula Lab, and SGK Inc.

View My Projects Get in Touch

Technologies & Tools

Python

Java

AWS

SQL

C++

TensorFlow

React.js

Django

Git

MongoDB

My Projects

Mosaic Ideation Hub

End-to-end ad ideation web app utilizing an AWS Bedrock pipeline to automate ad generation based on brand inputs. Integrated generative AI models to create high-quality visuals, enabling real-time image outpainting and dynamic text overlay using a customizable UI. Built a full-stack system with a Flask backend and a React frontend, leveraging AWS Lambda for serverless processing and S3 for image storage.

Python
React.js
AWS Bedrock
Flask
AWS Lambda
AWS S3
HTML/CSS

Image QC Analyzer

This Flask-based Image QC application analyzes image attributes and text/font properties to ensure compliance with predefined standards. It extracts resolution, DPI, color profile, margins, and background type using Pillow, while OpenCV and pytesseract handle OCR-based text detection and font size estimation. The app compares extracted attributes against predefined client rulesets and highlights mismatches. A simple web interface allows users to upload images, view extracted data, and check compliance in real-time.

Python
Flask
OpenCV
pytesseract
HTML/CSS
JavaScript

Chest X-Ray Abnormality Predictor

Create and trained a deep learning model to predict multiple (multilabel) abnormalities and diagnoses from 14 different categories. Applied transfer learning with ResNet50 architecture to enhance the performance of my classifier, leveraging pre-trained weights on the CheXpert dataset for improved accuracy in irregularity diagnosis. Plotted loss per epoch to analyze model accuracy during training. Model was used to help radiologists at MD Anderson Cancer Research Center.

TensorFlow
PyTorch
Python
NumPy
Matplotlib

Database Analyzer for Asset Valuation

Developed a web application accessing and displaying information from NoSQL database Cassandra, aiding in the centralization of asset valuation. Analyzed and displayed 50K+ transactions from Cassandra database using Java, Spring Boot, and Node.js. Developed a UI using React.js, JavaScript, and HTML to visualize database findings.

Java
SQL
SpringBoot
React.js
Node.js