About Me
I'm a computer science student at Pennsylvania State University with a deep fascination for the theoretical foundations and future potential of technology. My academic journey is guided by a persistent curiosity about how we can push the boundaries of computation — particularly through the lens of quantum computing and algorithm design.
I believe that breakthroughs in algorithmic thinking can power the next generation of responsible and scalable systems. My current interests lie at the intersection of quantum-safe cryptography, quantum algorithms for computer vision, responsible AI and LLM evaluation, computational biology, and building elegant, efficient solutions to complex computational problems.
Whether I'm optimizing classical systems or exploring the capabilities of quantum frameworks, I'm driven by a desire to bridge theory with application, ensuring that the technologies of tomorrow are not only innovative, but also ethical and impactful.
Education
Pennsylvania State University
Aug. 2023 – May 2027
B.S. Computer Science · GPA 3.90/4.0 · University Park, PA
Relevant Coursework
Skills
Languages
AI / ML
Web & Systems
Cloud & DevOps
Quantum
Data & Tools
Experience
Summer Undergraduate Research Fellow (SURF) – Purdue CGVLAB
Summer 2026 · West Lafayette, IN
- Researched the intersection of quantum computing and computer vision under Prof. Daniel Aliaga and Prof. Eric Samperton.
- Formulated image instance segmentation as a merge-only QUBO over region adjacency graph edges, converted it to an Ising Hamiltonian, and solved it with three interchangeable optimizer backends sharing one problem-construction and decoding API.
- Executed QAOA on a 156-qubit IBM Heron QPU, recovering the exact brute-force optimum on 672/672 validated subgraphs and reaching a 0.784 approximation ratio on hardware vs. 0.826 on a noiseless simulator.
- Used classical parameter transfer to collapse a queue-bound, multi-job workflow into a single hardware submission with no measurable quality loss.
Software Engineering Intern – Jubilant Pharmova
Summer 2025 · New Delhi
- Worked in a team of 10 developers to optimize workflow across 46 radiopharmacies in the US, reducing ingredient waste by 15%.
- Architected a LangChain AI agent integrating CPM tracking, JSON, and SQL Server sources through structured tool-calling to autonomously extract and compute pharmacy report metrics, eliminating manual review and cutting data processing time by 90%.
- Researched AI/ML automation tools to enhance radiopharmaceutical logistics while ensuring HIPAA compliance.
Undergraduate Researcher – Blockchain Data Intelligence Lab
Spring 2025 - Present · Pennsylvania State University
- Researching secure decentralized quantum computing and its intersection with cross-chain crypto bridges under Prof. Madhusudan Singh.
- Co-authored QLink, a quantum-safe Layer 3 bridge architecture combining post-quantum cryptography, quantum key distribution, and hardware security modules.
Research Assistant at the RAISE Lab – Ethical use of AI
Fall 2024 - Present · Pennsylvania State University
- Doing research in The RAISE Lab (Responsible AI for Social Emancipation), under Professor Amulya Yadav, ensuring Fairness, Accountability, and Transparency (FAccT) in machine learning.
- Designed and ran an evaluation framework benchmarking four frontier LLMs on 176 rare-disease clinical vignettes drawn from House M.D. case studies; no model exceeded 38.6% diagnostic accuracy. Published as an AAAI 2026 workshop paper.
- Built a detection pipeline identifying copyrighted material present in LLM pretraining corpora, surfacing data-provenance risks relevant to responsible model training.
- Bridging the gap between AI researchers and non-profit organizations through human-computer interaction.
Software Engineering Intern – Ministry of Electronics & IT
Summer 2024 · Government of India
Represented the Ministry in providing technical assistance to startups. Built a decentralized voting system using blockchain to ensure transparency, immutability, and trust in elections. Conducted 10+ blockchain workshops impacting ~100 attendees.
Machine Learning Intern – Agritech & Public Health
Summer 2024 · Grus N Grade Pvt. Ltd.
Designed ML models to enhance crop choice and farming efficiency for Indian farmers. Built a web crawler aggregating 50+ crop profiles across Indian states—focusing on nutrients, soil type, water needs, and texture—to support predictive modeling.
Math Proctor & Grader
Fall 2024 - Present · Pennsylvania State University
5-star rated proctor with 50+ sessions. Recognized by professors for reliability and integrity in exam supervision. Graded weekly assignments for over 100 students, ensuring fairness and consistency across multiple math courses.
Treasurer – Hack @ Penn State
Fall 2024 - Present · Pennsylvania State University
Managing budgets, reimbursements, and UPAC funding proposals; maintaining transparent financial records and ensured timely, compliant payments to support large-scale hackathon operations.
Publications
Do Personalized Evaluation Functions Reflect Human Preferences? A Study of Weighted Proximity in Algorithmic Recourse
Authors: Arsh Gupta, et al.
Algorithmic recourse tells a person what to change to flip an unfavorable automated decision, and most methods rank those suggestions with a weighted proximity function meant to stand in for what the person actually cares about. This work studies whether such personalized evaluation functions really track human preferences, and what that means for how recourse is evaluated.
Authors: Arsh Gupta, Ajay Narayanan Sridhar, Bonam Mingole, Amulya Yadav
This study introduces a dataset of 176 symptom-diagnosis pairs from the "House M.D." television series, validated for teaching rare disease recognition. It evaluates four state-of-the-art large language models (LLMs) on narrative-based diagnostic reasoning tasks, revealing significant performance variations — no model exceeded 38.6% diagnostic accuracy — and highlighting challenges in rare disease diagnosis.
Authors: Joao Vitor Barros Da Silva, Arsh Gupta, Madhusudan Singh
QLink introduces a quantum-safe Layer 3 interoperability protocol for secure cross-chain communication in blockchain networks. It integrates post-quantum cryptography, quantum key distribution, and hardware security modules to protect against both classical and quantum threats, providing a scalable and decentralized foundation for blockchain interoperability.
arXiv:2512.18488Outlier Detection in Single-Cell Transcriptomics Reveals Disease-Enriched Cytotoxic Immune Populations
Authors: D. Gupta, Arsh Gupta, et al.
Applies outlier detection to single-cell transcriptomic data to surface rare cell populations that standard clustering tends to average away, identifying cytotoxic immune populations that are enriched in disease samples.
Preprint
Projects
Next.js, FastAPI, PostgreSQL, Claude API, MediaPipe
- Built a full-stack training platform with an agentic Claude API coach that autonomously calls seven user-scoped tools (training history, send logs, grade pyramid, video analyses, detected weaknesses) to answer questions, streaming responses to the client over Server-Sent Events.
- Designed the schema and access layer so every tool call is scoped to the requesting user, preventing cross-account data leakage across 9 tables and 35 authenticated endpoints.
- Built a MediaPipe Pose computer-vision pipeline scoring climbing-attempt video on four movement fundamentals with severity-ranked coaching feedback, processing a 20-second 1080p clip in ~4 seconds on an Apple M3 Pro.
Python, PennyLane, PySCF, RDKit
- Built an end-to-end VQE pipeline estimating SARS-CoV-2 main-protease inhibitor binding energy, covering ligand preparation, active-space reduction, and energy estimation.
- Implemented a from-scratch UCCSD-style ansatz with parameter-shift gradients and multi-restart initialization, validated against exact FCI references.
- A full-stack web application using Node.js allowing employees to anonymously rate and review their managers.
- Implemented a machine learning model to analyze user comments and encrypted API endpoints to protect user anonymity and data integrity.
- Created a feature for users to add or select tags; backend supports high-volume data traffic.
HackPSU: AI Glaucoma Detector | Fall 2023
- Led a team of 5 to develop an AI model that detects glaucoma with 95% accuracy.
- Used Python, Torch, and StyleGAN3 to process and analyze over 120K retinal images.
Achievements
Academic Excellence
GPA: 3.90 / 4.0
🎓 Honored with Dean's List: Spring 2024, Fall 2024, Spring 2025
Research Honors & Service
- 🔬 Selected as a Summer Undergraduate Research Fellow (SURF) at Purdue University CGVLAB, researching quantum optimization for computer vision on IBM Heron hardware.
- 📝 Program Committee Member & Reviewer — AAAI/ACM Conference on AI, Ethics, and Society (AIES) 2026.
Certifications
✅ Completed Harvard CS50 AI Certification, Google AI Essentials, and Hedera Blockchain Developer Certification — all with 100% grades.
Virtual Job Simulations (Forage)
Goldman Sachs – Software Engineering
Verizon – Cloud Platform Job Simulation
Accenture UK – Developer
J.P. Morgan – Software Engineering
Skyscanner – Front-End Engineering
Get in Touch
Feel free to reach out via email, or find me on LinkedIn, GitHub, and Google Scholar.