About
I got into programming in high school building chatbots and automation tools, and that curiosity quickly turned into a drive to build software people actually use. At UT Dallas, I've expanded into full stack, machine learning, and cloud engineering, building production systems like a React Native app serving 300+ users at NRVE and ACM's event platform that onboarded hundreds.
I've also worked on physics informed ML models in the Reliability and Design Automation Lab and apply my CE background through Formula Racing, contributing to powertrain and system design. Outside the classroom, I'm a Marketing Lead for HackUTD and run Sprout Designs, where I've built digital experiences and branding for clients like OpTic Gaming, FaZe Clan, and the Miami Dolphins.
Across everything I build, from ASL to speech to AI systems, I focus on creating fast, scalable solutions that actually matter.
Experience
Fannie Mae · Dallas, TX
Incoming Software Engineering Intern
NRVE · Dallas, TX
Software Engineering Intern
- Built a personalized education app using React Native, Firebase, and Python, serving 300+ beta users.
- Designed and implemented cloud infrastructure using AWS SageMaker, AWS Lambda, and Azure Web Apps, reducing provisioning time by 30% while maintaining scalability.
- Established CI/CD pipelines using GitHub Actions with automated unit and integration testing, improving release efficiency across multiple Agile sprint cycles.
- Containerized backend services using Docker and deployed on Kubernetes, enabling scalable model inference and zero-downtime deployments across environments.
Association for Computing Machinery – Outreach · Dallas, TX
Software Development Lead
- Architected workflow-driven systems using React, TypeScript, Firebase, and Spring Boot.
- Designed JSON-based data schemas to power dynamic UI forms and application state management.
- Led Agile (Scrum) ceremonies including backlog grooming, sprint planning, and daily stand-ups using Jira.
University of Texas at Dallas · Dallas, TX
Machine Learning Research Assistant – Reliability & Design Automation Lab
- Engineered and deployed three ML pipelines in Python, TensorFlow, PyTorch, and Scikit-learn for battery life prediction with 10,000+ data points.
- Built data processing and training components with REST API endpoints, reducing reconfiguration time by 60%.
- Improved prediction confidence by 20% using conformal prediction and synthetic data generation.
Mark Cuban Foundation · Dallas, TX
Technical Ambassador & AI Program Lead
- Built and deployed an AI-powered computer vision web app using Python, Flask, and Azure services, scaling to 1,500+ users.
- Led hands-on machine learning sessions for 1,500+ students, improving engagement by 20%.
- Led data-driven outreach using Python-based analysis and SQL reporting to streamline student workflows.
Philosophy
Define the use case
Start with the product problem, user context, and a clear outcome worth building for.
Shape the interaction
Turn the idea into flows, interface structure, and prototypes that can be tested early.
Build the system
Engineer reliable frontends, services, and infrastructure that support real-world use.
Refine the details
Polish motion, performance, and interface edges so the product feels intentional.
Work
R.A.C.E.
Telemetry + AI Tooling
Race-engineering software that turns telemetry, radio context, and strategy rules into a fast decision surface for live sessions.
Vertera
Analytics Platform
A business analytics product built around customizable dashboards, forecasting, and clear reporting for small teams.
SeatSwap
Travel Experience Concept
A flight seat-swapping experience designed for real-time availability, request flows, and quick passenger coordination.
ASL Translator
Computer Vision · Edge AI
Real-time ASL-to-speech translator using MediaPipe landmarks and a custom gesture model deployed on Raspberry Pi via TensorFlow Lite, achieving 92% accuracy on the ASL alphabet.
FinSights
RAG · Fintech
Cross-platform React Native app with a RAG pipeline using LangChain, FAISS vector search, and Sentence-Transformer embeddings for querying financial documents.
HackAI Website
Hackathon · Web
Marketing site for HackAI, a student-run AI hackathon at UTD. Teams go from idea to demo in 24 hours through workshops, mentorship, and hands-on building.
MonteCore
Quantitative Trading Platform
Full-stack quant platform that backtests moving average strategies on historical equity data, evaluates risk-adjusted performance, and runs progressive Monte Carlo simulations to model uncertainty.
Stack
Contact
Open to internships, product teams, and ambitious software work. If you are building something thoughtful and need a developer who cares about both craft and execution, let's talk.
dhivyeshrathi@gmail.com