Noah Arooji

EducationClick to focus

Expected May 2027

Bachelor of Science Computer Science, Bachelor of Arts Applied Statistics

GPA: 3.80 / 4.00

Relevant Coursework: Data Structures and Algorithms, Machine Learning, Artificial Intelligence, Regression Analysis

Languages:Java, C, Python, JavaScript/TypeScript, R, SQL, HTML, CSS, MATLAB

Frameworks:React.js, Next.js, Node.js, TailwindCSS, NumPy, PyTorch, Pandas, REST APIs, OOP, Mockito, Git, Agile

Certifications:2025 Oracle Cloud Infrastructure (OCI) Foundations, 2025 Oracle AI Foundations Associate, CITI Researcher

May 2026 – August 2026

Software Development Engineer Intern

Herndon, VA

Designed and built a value insights system surfacing $15M in actionable price intelligence daily, replacing manual spreadsheets with an on-demand dashboard and cutting manual record pricing by 99% for internal metering teams
Architected a distributed pricing application in Java hosted on AWS Fargate, leveraging an event-driven pipeline to process 7,000 billing records per minute with 99.94% file success rate
Engineered data ingestion to handle 5M writes per hour to InfluxDB, optimizing batch writes and time-series indexing to serve sub-second time-range queries to React.js (TypeScript) dashboards

May 2025 – August 2025

Software Engineer Intern

Virginia Beach, VA

Led a cross-functional team of 5–10 through development of a full-stack client prospecting tool surfacing 20+ high-quality business leads and $1M in generated business pipelines per use
Built interactive data visualization components in React.js with modular state management, rendering client metrics across filterable lead scoring views, reducing lead evaluation time by 40% across 3 internal teams
Architected data ingestion and enrichment with REST API endpoints leveraging Express.js and Oracle ADW to fetch 50,000+ records in sub 600ms latency

April 2025 – Present

Machine Learning Researcher

Charlottesville, VA

Trained a CLIP-based multimodal retrieval model using PyTorch, aligning time-series chart embeddings with natural language descriptions across 3,000+ Federal Reserve economic blog posts
Built scalable preprocessing pipelines in Pandas and NumPy, transforming raw economic time-series records into structured contrastive learning pairs, reducing data preparation time by 95%
Optimized model training through systematic data quality filtering and time-series alignment, removing 30+ insufficient data pairs and improving validation recall convergence by 45% across encoder configuration experiments

March 2025

React.js, TailwindCSS, Vite, Chrome API

Built a Chrome Extension using React.js, TailwindCSS, and Vite, reducing user ChatGPT token waste by 30% while visualizing users’ sustainable AI usage patterns
Implemented lightweight data persistence using the Chrome Storage API, locally caching token counts and prompt metadata for real-time efficiency tracking
Rendered dynamic data visualization modules with Recharts, transforming token-level logs into clear trend insights on user efficiency and sustainability

November 2024

Expo, React.js, Flask, Google Vision OCR, Amazon AWS

Engineered a full-stack macronutrient tracking application using React.js, Flask, and Expo, supporting cross-platform deployment for web and mobile clients
Deployed and managed the application on AWS EC2 with S3 storage, integrating RESTful Flask APIs for real-time nutrient computation and user authentication
Scaled backend services to reliably handle 100+ active users, improving request latency and data synchronization across devices

January 2025 – August 2026

Personal Trainer

Charlottesville, VA

Guided over 20 clients through 100+ personalized sessions, breaking down complex goals into actionable steps for long-term client growth
Constructed measurable performance metrics, increasing program adherence and goal completion by 25%