cv

Graduate Student Researcher, Biomedical Informatics

Arizona State University — SHARES Lab

Research on privacy-preserving clinical decision support using FHIR and CQL. Built SHARES Consent Engine for granular sensitive data segmentation; developed clinician-centered FHIR visualization tools and conducted synthetic FHIR cohort evaluations.

• FHIR R5 Consent evaluation with CQL
• Granular sensitivity labeling & consent thresholds
• Clinician-focused FHIR visualization (D3.js/TypeScript)
• HAPI-FHIR server + terminology integration

Exempt Non-Officer - Software Development, Investment Banking

Credit Suisse Services AG

• Developed and managed a global order management system for Stocks, Futures, and Options within the Cash Equity business, ensuring precise execution to meet diverse client needs amid fluctuating market conditions. • Managed deployments and modifications of trading services encompassing short sell order locators, compliance and administrative applications. • Oversaw 85+ production level changes (RFCs) over two years, demonstrating expertise in C# and WinForms development along with database skills in Sybase and Oracle. • Recognized with 2 RAVE (Recognizing Value and Excellence) awards for exceptional partnership, accountability, and contributions to project success

• Software Development
• C#
• Order Management Systems
• Database Management

Technical Analyst, Investment Banking

Credit Suisse Business Analytics

• Created and revamped crucial daily ETL (Extract, Transform, and Load) jobs to fetch data from stock exchanges to the Credit Suisse Program Trading database, incorporated by traders to book and review trades. • Interacted and coordinated with stakeholders across regions (US, UK, APAC) to discuss software requirements,design, and creation of automated reports based on C#’s .NET Core and the .NET Framework. • Reduced ETL job creation time by 70% by templating and revamping legacy methods for the same.

• Data Analysis
• ETL
• SQL
• C#

Technology Intern, International Wealth Management

Credit Suisse

• Collaborated on delivering Continuous Integration and DevOps solutions for the IWM Tech Department. • Engaged in the deployment of Load balancers and OpenShift Clusters for cross-functional projects

• Openshift
• C#

Research Assistant, Digital Healthcare

Visvesvaraya National Institute of Technology, Nagpur

• Conducted quantitative research towards the creation of a novel multimodal early detection approach for mental health ailments based on an end-to-end machine learning pipeline, under the guidance of Dr. Praveen Kumar. • Designed remote sensing protocols for the collection of user smartphone data (via AWS S3 Buckets) and extracted high-level features for smartphone, audio, and visual data (using NumPy and Pandas) using Python. • Optimized memory efficiency of data collection by 90% over the pre-existing methodology. • Co-authored 2 internationally published papers on the topic, providing valuable insights into the approach’s effectiveness for future researchers

• Quantitative Research
• Multimodal Machine Learning
• Digital Healthcare
• Python

Software Developer, Google Summer of Code 2019

The Apache Software Foundation

• Researched and implemented pseudorandom number generators for the Apache Commons open-source project. • Accomplished the integration of LCG, Permuted Congruential Generator variants, and other generators into the commons-rng repositories, adopted by thousands of developers globally. • Applied Java’s Maven, JUnit, and JMH for project building and test suite creations for all 3 variants of generators

• PRNG
• Java

Internship Trainee, Central Technical Services

Reliance Infrastructure (now Adani)

• Collaborated on delivering Continuous Integration and DevOps solutions for the IWM Tech Department. • Engaged in the deployment of Load balancers and OpenShift Clusters for cross-functional projects

• JavaScript
• .NET Framework
• Web Development

Peer Reviewer

International Journal of Human-Computer Studies (Springer Nature)

Reviewed manuscripts for IJHCS.

• Peer Review

Peer Reviewer

NeurIPS Workshop on Time Series for Health (TS4H)

Reviewed submissions for TS4H workshop.

• Peer Review

Elemental Member

American Association for the Advancement of Science (AAAS) — Neuroscience

AAAS membership in Neuroscience section.

• Membership

Student Member

American Medical Informatics Association (AMIA)

Professional student membership in AMIA.

• Membership

Career Guidance Outreach

Antarang Foundation

Led career guidance initiatives for at-risk youth in technology careers.

• Outreach
• Education

STEM Mentorship & Diversity Advocacy

Credit Suisse

Co-founded a mentorship program to empower underrepresented STEM professionals.

• Mentorship
• Diversity
• Community

Technical Advocacy & Content

Credit Suisse

Led engagement for global coding challenges and edited the TA newsletter.

• Content
• Advocacy

MS (Thesis)

Arizona State University, United States of America

Computer Science (Biomedical Informatics)

Executive PG Program

International Institute of Information Technology, Bangalore

Data Science

Bachelors (B.Tech)

Visvesvaraya National Institute of Technology, Nagpur

Computer Science and Engineering

• Data Structures and Program Design
• Software Engineering
• Operating Systems
• Neuro-Fuzzy Techniques
• Database Management Systems
• Introduction to Object Oriented Methodology
• Design and Analysis of Algorithms
• Theory of Computation

Herbold Foundation Scholar

Herbold Foundation

Awarded to master's students in Computer Science with high academic and research potential.

NIH-Funded Biomedical Informatics Researcher

National Institutes of Health

Supported research in privacy-preserving clinical decision support.

RAVE Award

Credit Suisse

Recognizing Value and Excellence for partnership and accountability.

Graduate Student Resilience Award

ASU Graduate Student Government

Recognized for resilience and contributions to the graduate community.

Ryan Star Award for Academic Excellence

Ryan Group of Institutions

Awarded for outstanding academic performance.

FHIR-Based Visual Simulation of Consent-Driven Granular Data Segmentation

AMIA Annual Symposium (Systems Demonstration)

Demonstration of SHARES consent-driven segmentation and visualization for privacy-preserving CDS.

Multimodal Depression Detection: Using Fusion Strategies with Smart Phone Usage and Audio-visual Behavior

World Scientific

Detecting depression is complex due to variable symptoms arising from individual differences. Our research aims to create a novel classification system for diagnosing depression, considering historical data (like activity levels) and verbal/non-verbal cues (pitch, gaze). We built a real-world dataset combining 14-day smartphone records and audio-visual data, while extracting physiological/behavioral features. Using Decision trees and SVM classifiers with fusion methods, we found SVM with late fusion achieves 89% accuracy. We validated this approach on the DAIC-WOZ dataset, reinforcing its effectiveness.

A novel multi-modal depression detection approach based on mobile crowdsensing and task-based mechanisms

Springer

We propose a novel depression detection approach by integrating task-based and Mobile Crowd Sensing (MCS) methods. Our pipeline includes data collection, feature extraction, fusion, and classification. Experimental findings show combining multi-modal features excels, especially fusing all three modalities. SVMs achieve 86% accuracy. Our approach outperforms state-of-the-art techniques on benchmark data.

SHARES Consent Engine (FHIR + CQL)

Deterministic granular sensitive data segmentation using FHIR R5 Consent and CQL with CDS Hooks integration.

• Consent thresholds
• Sensitivity labeling
• HAPI-FHIR integration

FHIRLight: Clinician-Centered FHIR Visualization

Lightweight FHIR bundle loader and multipanel visualization (vitals, meds, labs) with consent preview.

• TypeScript
• D3.js
• FHIR R5

Granular Segmentation Benchmark on Synthetic FHIR Cohort

Evaluation on ~10k Synthea patients for throughput, precision, and cost of consent thresholds.

• Python
• Benchmarking
• FHIR R5

HAPI-FHIR Server + CQL Evaluation

HAPI-FHIR main and terminology servers to support Library/$evaluate workflows for consent rules.

• HAPI-FHIR
• CQL
• Terminology

SHARES-CLI

CLI utilities to encode CQL to base64, create FHIR bundles, and POST to HAPI-FHIR endpoints.

• CQL
• FHIR
• CLI

Sharpnr – AI-Powered Academic Assistant

Multi-agent system aggregating Canvas, Slack, and Calendar to deliver notifications and study plans.

• FastAPI
• React/Next.js
• MongoDB

MRI Style Transfer with CycleGAN

Generates T2-weighted MRI from T1 scans to augment data and support diagnostic accuracy.

• TensorFlow
• CycleGAN
• Medical Imaging

Indian Sign Language Translator

Orchestrated and contributed to the development of an Indian Sign Language Translator to aid individuals suffering from hearing and vocal impairments. The project satisfied the following criteria - Near-Real-Time Application, Background independence, and Illumination independence.

• Near-Real-Time Application
• Background independence
• Illumination independence

Genetic Algorithm Implementation

Created an implementation of genetic algorithms for generating solutions of the Travelling Salesman Problem for a large fully connected graph (about 50 nodes). This project involved making fitness, crossover, and randomized Euclidian distance functions with simulators for evolution and graphing accuracy.

• Fitness functions
• Crossover functions
• Randomized Euclidian distance functions

Comparison of Clustering Techniques

Applied clustering techniques like K-Means, BIRCH, and Agglomerative Clustering (AGNES) on the US News and World Reports College Data for the generation of insights into the accuracy of clustering methods when bifurcating Private and Public colleges.

• K-Means
• BIRCH
• Agglomerative Clustering

Feature Segmentation via Multi-Res-U-Net Masking

Researched an experimental implementation of MultiResUNet for facial feature segmentation using the CelebAMaskHQ dataset. The project aimed to utilize and discover the compatibility provided by MultiResUNet models initially used for Medical Image segmentation.

• MultiResUNet
• Facial Feature Segmentation
• CelebAMaskHQ dataset

Resource Allocation within a Multi-Client Server

Implemented a program for allocating resources that makes use of a single server and can deal with simultaneous demands from a number of client apps. The project incorporated a wait-list queue and locked the resources until they are released by the client apps to maintain resources in sync with all clients.

• Resource Allocation
• Multi-Client Server
• Wait-list Queue

Kerberos Protocol Implementation

User authentication and ticket generation were prepared using the Kerberos protocol between clients and servers to manage user accesses to services. A Key-Dependent Additive Cipher implementation was applied to encrypt the channels of communication between the client and the servers.

• Kerberos Protocol
• User Authentication
• Key-Dependent Additive Cipher

Command Shell Implementation

Built a command shell using OS system calls (fork, wait, exec, etc.) to execute built-in Linux commands. Contains the ability to support multiple commands in one prompt along with support for redirection, external signals, and error handling.

• OS System Calls
• Linux Commands
• Redirection

Logistic Regression Classifier for Movie Reviews

Created a model to categorize movie reviews (positive or negative) using the IMDB dataset. Utilized sci-kit learn for model creation and NLTK for feature extraction. Optimized model accuracy through hyperparameter tuning.

• Logistic Regression
• IMDB Dataset
• Hyperparameter Tuning