Pittsburgh, PA · Open to software engineering internships

Aryan Gupta

Mathematics and Computer Science @ Carnegie Mellon University

I love building things that solve problems I have and converting ideas into real projects!

View projectsContact me

About

Education, interests, and stuff I'm passionate about

Carnegie Mellon University

B.S. in Mathematical Sciences and Computer Science

2025 - 2028 (Expected)

0GPA

Princeton University

Dual enrollment in mathematics

MAT 214 · Numbers, Equations, and Proofs

South Brunswick High School

National Merit Scholarship Winner · Math Excellence Award

Selected coursework

21-270 Introduction to Mathematical Finance
21-241 Matrices and Linear Transformations
21-128 Mathematical Concepts and Proofs
15-150 Principles of Functional Programming
15-122 Principles of Imperative Computation
15-251 Great Ideas in Theoretical Computer Science
18-213 Introduction to Computer Systems
21-259 Calculus in Three Dimensions
21-260 Differential Equations

Interests

Hi! My interests are in building systems and projects that either solve problems I have or deepen my understanding of certain things I don't know much about. I think there's nothing not worth learning about, and I love exploring things I've never learned about before! I am most passionate about going deep in the technical side of projects and really understanding the underlying implementations.

Skills

Languages

PythonCC++JavaSQLSMLAssembly

Systems

LinuxMemory ManagementConcurrencyNetworkingProcessesSignalsDebugging

AI & Data

QwenPaddleOCRTesseractPandasNumPyJupyter

Web & Tools

ReactTailwind CSSGitGitHubGDBVS Code

Experience

May 2026 — August 2026

Software Engineering Intern

Mphasis AI · New York, NY

  • Built an end-to-end document-boundary detection pipeline combining Qwen vision-language models with Tesseract and PaddleOCR for large, multi-page PDFs.
  • Compared Qwen-only, Qwen + Tesseract, and Qwen + PaddleOCR pipelines to balance accuracy, recall, and processing speed.
  • Designed an OCR flagging path that routes only low-confidence pages to PaddleOCR.
  • Automated evaluation against labeled ground truth using accuracy, precision, recall, F1, and false-positive/false-negative analysis.
  • Optimized PDF rendering, OCR extraction, batch processing, and inference across hundreds of pages, with reproducible Jupyter pipelines and Excel reports.
PythonQwenTesseractPaddleOCRPandasJupyter

December 2023 — December 2024

Research Intern — Physics and Astronomy

Rowan University · Glassboro, NJ

  • Analyzed electrochemical datasets to quantify variability, evaluate signal stability, and identify trends across repeated trials.
  • Contributed to research on LaF₃ microsensors, WO₃ electrochromic films, and LiFePO₄ thin-film materials.
  • Synthesized findings from more than 20 publications and wrote a standard operating procedure for future researchers.
Data AnalysisElectrochemical MaterialsExperimental Research

Projects

Selected work

Jun 2026

Dynamic Memory Allocator (C)
  • Implemented a 64-bit dynamic memory allocator in C with explicit segregated free lists, block splitting, and coalescing to manage heap allocation and deallocation
  • Developed heap consistency checks and used GDB to identify memory corruption, segmentation faults, and violations of allocator invariants

Jul 2026

Tiny Shell (C)
  • Implemented a Unix shell in C supporting foreground and background job execution, job control, signal handling, and input/output redirection
  • Developed process management using fork, execve, waitpid, and POSIX signals to correctly manage concurrent processes and prevent race conditions during job execution

Feb 2026 — Mar 2026

Riseva (2nd Place in Carnegie Mellon Product Hackathon)
  • Built an AI-powered learning platform using Python and React that generated personalized learning experiences and captured student interaction data for analysis
  • Developed an analytics dashboard that transformed learning data into insights for tracking student mastery and identifying learning gaps

Nov 2025 — Dec 2025

Monte Carlo Option Pricer (Python)
  • Implemented Monte Carlo simulations to price European call options under geometric Brownian motion, including confidence interval estimation and statistical error analysis
  • Applied antithetic variates for variance reduction and analyzed simulation convergence against analytical Black-Scholes prices

Jul 2026

Concurrent Web Proxy (C)
  • Implemented a multithreaded HTTP proxy server in C using POSIX threads, socket programming, and synchronization to concurrently process client requests and forward traffic to origin servers
  • Developed a thread-safe in-memory cache with LRU eviction and reference counting to reduce repeated-request latency while preventing race conditions during concurrent access

Dec 2025 — Jan 2026

Moving Average Crossover Backtester (Python)
  • Developed an event-driven Python backtesting engine for quantitative trading strategies on SPY with next-day execution to eliminate look-ahead bias
  • Evaluated strategy performance across market periods using CAGR, Sharpe ratio, volatility, and maximum drawdown to analyze risk-adjusted returns

Contact

Emailaryangup@andrew.cmu.edu
GitHubgithub.com/a-gupta123
LinkedInlinkedin.com/in/aryangupta111