EXPERIENCERESEARCHPROJECTSENGAGEMENTCONTACT

WORK DOSSIER

TEJAS GUPTA

SOFTWARE ENGINEER

Experience

Software Engineer Intern

Magic Labs

Sept 2024 — Present

  • Enabled real-time pose evaluation for 200+ students with 95% pose classification accuracy by developing an AI-powered full-stack pose detection platform.
  • Reduced frontend latency by 40% across the application by architecting the user interface using the MVVM pattern in React and implementing real-time synchronization with Firebase Realtime Database and Firestore.
  • Protected user data across 1,000+ daily cloud transactions by implementing Firebase Authentication and Security Rules to enforce role-based access control.
  • Improved backend maintainability by implementing RESTful APIs using Clean Architecture, separating domain models, use-case services, and API controllers to enforce modular and testable service design.

Software Engineer Extern

Atlatl Advisers

Sept 2025 — Dec 2025

  • Enabled financial advisory services by building an ADA-compliant platform using Next.js.
  • Improved client onboarding by integrating an AI chatbot that guides users through financial planning workflows.
  • Increased user engagement by implementing 12 interactive financial calculators for portfolio and retirement planning.
  • Enabled non-technical content updates by integrating Strapi CMS for dynamic management of the site.

Software Engineer Intern

Ernst & Young

June 2025 — Aug 2025

  • Improved fraud detection accuracy by 28% across input tax credit records by developing an AI-based anomaly detection system to identify false invoices and tax mismatches.
  • Enabled reliable payment processing for financial transactions by implementing secure Stripe REST APIs with idempotent transaction handling and automated retry safeguards.
  • Strengthened cloud deployment security across AWS infrastructure by implementing compliant service configurations aligned with enterprise security and regulatory standards.

Software Engineer Intern

Vikas Cable Company

May 2023 — Aug 2023

  • Improved cable test batch traceability by developing a Java system that automated resistance and insulation tests.
  • Improved backend modularity by implementing OOP principles and Factory/DAO design patterns.
  • Enabled reliable test data storage and retrieval by designing SQL schemas and integrating them with JDBC.

Research

Sensing motion beyond sight.

Keyboard Interaction Sensing Through Opaque Structures Using mmWave

A graduate research project investigating whether subtle hand and finger motion can be inferred through obstructed environments using radar reflections, signal processing, and predictive modeling.

Jan 2026 — Present

Grad Research Project

Developed an FMCW radar processing pipeline to sense keyboard interaction through opaque structures using mmWave data.

Extracted micro-motion signatures from noisy reflections by building Python-based signal processing workflows for denoising, temporal alignment, and feature generation.

Improved keystroke prediction robustness by training machine learning models on radar-derived motion features across varied hand motion patterns.

Core Question

Can radar reflections capture enough structured micro-motion to distinguish discrete keyboard actions even when direct visual access is blocked?

Approach

Combine FMCW radar acquisition, feature engineering, and machine learning to transform weak motion traces into a more reliable interaction signal.

Potential Impact

This work could support privacy-aware interfaces, assistive systems, and sensing in environments where camera-based interaction is impractical or undesired.

Experimental Study

Selected Works

Projects

March Madness Betting Platform preview 1March Madness Betting Platform preview 2March Madness Betting Platform preview 3March Madness Betting Platform preview 4

March Madness Betting Platform

A full-stack betting platform designed for real-time NCAA tournament interaction, combining a responsive frontend, Java-based backend services, automated sports data ingestion, and containerized deployment.

Enabled real-time NCAA tournament betting by building a full-stack platform using React, Java services, and MySQL.

Implemented backend service communication and networking logic across containerized components.

Automated sports data ingestion by developing a web scraper that populates MySQL with real-time updates.

Simplified deployment by orchestrating frontend, backend, database, and admin services using Docker Compose.

This project is not live at the moment, but the codebase is available on GitHub.

ReactJavaMySQLDocker
Not live right nowGitHub

Coming Very Soon

A new build currently in development. The concept, visuals, and launch details will be revealed very soon.

A new project is currently being designed and developed.

The interface, interaction system, and final scope are still being refined.

A full public preview and supporting details will be added once the build is ready.

More information, visuals, and links will be added as soon as the project is ready to be shared.

In ProgressDesignDevelopment
Coming very soon

Engagement

Learning in
motion.

Outside the classroom, I like staying close to technical communities and practicing regularly through clubs, security challenges, and problem solving.

Leadership

Programming Officer

Aviation Club · UW–Madison

Contribute to programming and technical initiatives within the club, helping shape events and activities that keep members engaged with aviation, engineering, and hands-on learning.

Community

Member

Cybersecurity Club · UW–Madison

Stay involved with the campus cybersecurity community through club participation, technical discussions, and exposure to applied security concepts and problem solving.

Practice

CTFs and LeetCode

Pwn College · Personal Learning

Regularly practice capture-the-flag challenges from Pwn College and solve LeetCode problems as a hobby and as a way to sharpen analytical thinking, debugging ability, and technical fluency.

Contact

Think I would be a great fit for your team?

I enjoy building thoughtful systems across product, full-stack engineering, and intelligent interfaces. Whether you are hiring, collaborating, or just want to connect, I would love to hear from you.

I usually reply quickly