Hello, I'm Sejal ッ Crafting data into insights, innovation, and impact, one algorithm at a time.

Author

I'm a software developer and data science enthusiast who thrives on transforming innovative ideas into tangible solutions. My journey is marked by countless hours dedicated to finding solutions, and my enthusiasm for coding, particularly in Python, machine learning, and engaging in impactful research projects. From deriving insights to crafting AI for social good, I thrive in environments where innovation and purpose intersect. I'm currently on the lookout for new opportunities to contribute, innovate, and grow. Let's connect and explore how we can create impactful stories together.

Analyzing

In a wide range of subject areas, I have analyzed structured and unstructured data to extract actionable business insights. I love to craft stunning and clever visualizations that illustrate surprising results.

Developing

I'm all about taking those smart machine learning models out of their Jupyter Notebook homes and dreaming up ways to make them do real, cool stuff in the real world.

Exploring

Off-screen, you'll catch me binge-watching K-dramas or admiring picturesque skies. Cats are my ultimate mood lifters. Currently watching Queen of Tears. HMU for some good recs :p

Professional Experience

Graduate Research Assistant

June 2023 – December 2023

University at Buffalo

Coordinated with 5-member team in applying GNNs for hospital monitoring, resulting in a 27.4% enhancement in patient movement prediction accuracy and optimizing resource allocation Adopted NLP techniques to process text-based comments, achieving patient outcome prediction accuracy of 89%

Programmer Analyst

January 2020 - July 2022

Cognizant

Performed exploratory data analysis using Tableau, Jupyter Notebook, and Python (Pandas, NumPy, and Bokeh) to craft 10+ insightful, interactive visualizations, facilitating better decision-making and client understanding Deployed a standalone pipeline through CI/CD to build and integrate test microservices with Docker, AWS, and CDK Managed and optimized database operations for over 8 million records in PostgreSQL and Oracle

Education

Master's Degree

August 2022 - Feb 2024

University at Buffalo

MS in Data Science

Bachelor's Degree

August 2017 - June 2020

Vishwakarma Institute of Information Technology

BE in Computer Engineering

Featured Projects

Dehazing image

AcademiQ AI

This project introduces an innovative approach to image dehazing using a deep learning model that leverages convolutional layers, residual connections, and concatenation techniques. Implemented with TensorFlow 2.0, this model demonstrates significant improvements over traditional state-of-the-art methods.

Dehazing image

Satellite Image Dehazing

This project introduces an innovative approach to image dehazing using a deep learning model that leverages convolutional layers, residual connections, and concatenation techniques. Implemented with TensorFlow 2.0, this model demonstrates significant improvements over traditional state-of-the-art methods.

Dehazing image

360° Object Detection & Assistance for Visually Impaired People

The project is centered around developing an AI-driven robotic assistant with 360-degree object detection and voice interaction, offering visually impaired individuals accurate navigation and object detection experience, leveraging 360°cameras and ResNet18 for comprehensive object detection.

Dehazing image

Q-learning Optimization for Grid-based Reinforcement Learning

This project explores the application of Q-learning and SARSA in a controlled grid environment, where an autonomous agent navigates a 4x4 grid to maximize rewards by collecting batteries and avoiding obstacles. Used Python, Gynamsium, Google Colab and Reinforcement Learning Concepts

Dehazing image

AlexNet from Scratch on SVHN Dataset

This project develops an AlexNet model from scratch to classify images using Pytorch, Kaggle Notebook and Python. It explores various optimization techniques and data augmentation methods to enhance model accuracy and generalization, achieving significant improvements in image categorization

Dehazing image

Time-To-Event Analysis in Clinical Trials

This R project conducts a detailed survival analysis of ovarian carcinoma patients, focusing on evaluating the efficacy of two different treatment protocols. Utilizing the Kaplan-Meier estimator and log-rank tests, it compares survival outcomes among patients