Hi, I'm Dev Sheth

Software Engineering Intern & Full-Stack Developer

I build innovative web applications and passionate about cloud computing, DevOps, and product management. Currently working at Fidelity Investments and pursuing Computer Science at UT Dallas.

Profile picture of Dev Sheth

Me

About Me

Who I Am

I'm a passionate Computer Science student at the University of Texas at Dallas, currently working as a Software Engineering Intern at Fidelity Investments. I love combining my technical skills with real-world problem solving, whether it's building web applications, designing beautiful user interfaces, or improving internal systems.

My journey in tech started with curiosity and has evolved into a deep passion for cloud computing, DevOps, product management, and UI/UX. I enjoy using tools like Figma to create mockups and collaborating on design-focused projects. I believe in writing clean, maintainable code and building products that make a real difference in people's lives.

What I Love

Cloud Computing
Traveling
Anime
Team Collaboration

2+

Years Experience

7+

Projects Completed

10+

Technologies

Education

University of Texas at Dallas

Richardson, TX

Bachelor of Science in Computer Science

Minor: Business Analytics and Artificial Intelligence

May 2027

Experience

Fidelity Investments

Software Engineering Intern

June 2025 - August 2025
Impact

Migrated a Node.js AWS Lambda from static configs to a centralized config store, enabling runtime config changes without redeploys and improving flexibility, maintainability, and operational efficiency.

Responsibilities
  • Integrated an internal config-store module for on‑demand configuration retrieval in a production Lambda service.
  • Implemented environment‑driven resolution of profile/label/keys; staged resolved configs in ephemeral storage for runtime use.
  • Leveraged a known‑good reference implementation to guide architecture and close parity gaps.
CI/CD & Monitoring
  • Maintained observability with Datadog metrics/logs to meet monitoring requirements and validate behavior post‑migration.
  • Used Jenkins to operate Lambda pipelines end‑to‑end; integrated Terraform changes with code updates for automated deployments.
  • Verified deployments across test/stage via logs, runtime configs, and system behavior checks.
Stack
Node.js AWS Lambda Terraform Jenkins Datadog Git / GitHub Enterprise
Highlights
  • Learned Lambda cold starts and ephemeral storage patterns, env‑var strategies, and centralized config across environments.
  • Partnered with DevOps and platform engineering to align on architecture and best practices.
  • Presented outcomes to senior engineering leadership: reduced deployment time, improved reliability, fewer config errors.

Projects & Research

First Project

AnimeTracker ❤️

A passion project I started in late November 2023 combining my love for anime with full-stack development. Users can create accounts, make lists ranking their favorite anime, and post opinions on episodes or topics. Built as a single-page application with React.js and styled using Tailwind CSS, it fetches anime data via the Jikan API and provides an interactive, modern interface.

The backend is powered by Node.js with Express.js handling API routes and PostgreSQL storing user-specific data. Authentication uses JSON Web Tokens to ensure each user has personalized lists and posts. Future plans include implementing AWS Amplify/S3 for media storage and AWS RDS/DynamoDB for cloud database scalability.

React.js Node.js Express.js Tailwind CSS PostgreSQL Jikan API JWT Auth Full-Stack Cloud Ready

FinTrack (ACM Research)

Built and trained a hybrid CNN-LSTM model in TensorFlow to predict S&P 500 movements using minute-level stock data, incorporating technical indicators to capture market behavior. Developed a trading strategy using high-confidence predictions, achieving an 18% return versus a 9% S&P 500 baseline. Final model classified movements as Up/Down/Flat with 62.5% accuracy across 5,400+ model-driven trades; visualized trends with Matplotlib to evaluate performance over time.

Python TensorFlow Keras Pandas Matplotlib

Hope Restored Inventory Tracking Software

Built a user-friendly inventory management system supporting 300+ unhoused residents. Started from Figma mockups and built a React.js front-end (CoreUI) with 5+ interactive pages. Implemented a dynamic inventory table with filtering, searching, and manual input that updates in real time to manage 500+ donated items. Added QR scanning in a Django-based backend to efficiently add or remove items in the PostgreSQL database.

JavaScript React.js Python Django PostgreSQL

Vib.IO (Spotif.ai)

A music recommendation system analyzing Spotify listening data using audio feature extraction. Built personalized song suggestions based on musical characteristics and listening patterns, leveraging Python audio tooling and lightweight app frameworks.

Python Librosa Gradio Streamlit Spotify API

TripPlanner

A carbon tracking website for airplane travel with interactive map displaying 15,000+ U.S. airport locations. Integrates public APIs for carbon footprint calculations and provides accurate environmental impact assessments.

React.js Python Flask APIs

Skills & Technologies

Languages

Java
JavaScript
Python
C/C++
SQL
HTML/CSS

Frameworks & Libraries

React.js
Vue.js
Next.js
Node.js
Express
Flask
Django
Pandas
TensorFlow
Bootstrap
Tailwind CSS
Bulma
Matplotlib
NumPy

Tools & Platforms

AWS Tools
Git
GitHub Enterprise
Docker
Terraform
Jenkins
Datadog
PostgreSQL
Figma
UI/UX Design

Get In Touch

Let's Connect!

I'm always interested in new opportunities, exciting projects, and connecting with fellow developers. Whether you have a question about my work or just want to say hi, feel free to reach out!