Biomedical engineering and machine learning. I build physics-informed models of blood flow, EEG-based wearables, and computational tools for drug discovery.

Currently at Stanford University (B.S. Electrical Engineering, B.S. Bioengineering, Class of 2029).

• Now — Computational Biology Intern, The MITRE Corporation (Advancing Defense Pharmaceuticals)
• Previously — Medical AI Researcher, Stanford University School of Medicine, Department of Radiology
• Previously — Research Intern, Harvard Medical School (hemato-oncology and immunology labs)
• Also — International Youth Coordinator, Jaipur Foot USA (BMVSS), the world's largest prosthetic-limb NGO — 2.3M+ beneficiaries across 42 countries
• Founded — Alliance Tutors, an international STEM tutoring venture (2020–2025)

• vascupinn — Physics-informed neural network solving Navier–Stokes with Carreau–Yasuda non-Newtonian rheology for blood flow in patient-specific vasculature. Covered by a USPTO provisional patent.
• nebulab-eeg — Single-channel EEG acquisition chain, overnight recording, and ESP32-S3 escalating-haptic wake firmware for a closed-loop sleep-inertia system.
• pinn-pulsatile-drug-delivery · pinn-aneurysmal-bifurcated · pinn-hybrid-experimental — Method components and case definitions for the three publications below. Each repo's Status section states plainly what is and isn't released.
• safesip — Drink-spiking detection device: spectroscopic sensing concept, whitepaper, and business case.
• newweb — Personal site in Next.js over a hand-coded generative illustration; seeded PRNG for hydration-safe static prerender, contrast verified by computation.

Sole author on all three.

1. PINN-Based Pulsatile Flow Modelling and Targeted Drug Delivery Optimization in Computational Hemodynamics — Journal of Pharmacy and Bioallied Sciences (Wolters Kluwer), December 2025
2. Hybrid PINNs with Experimental Data for Real-Time Hemodynamic Prediction and Clinical Decision Support — International Journal of Medical and Pharmaceutical Research, December 2025
3. Functional and Ethical Modeling of Non-Newtonian Hemodynamics in Aneurysmal and Bifurcated Vessels Using Physics-Informed Neural Networks — Scientific Culture, October 2025

• USPTO Provisional — "VascuPINN" blood flow modeling system (February 2025). Sole inventor.
• USPTO Provisional — A closed-loop multimodal stimulation system and method for reduction of sleep inertia (2026).

• Morehead-Cain Scholarship — full-ride merit scholarship to UNC Chapel Hill; declined for Stanford
• MITRE AI Hackathon — Winner (2026); the only high school student competing against graduate teams
• Special Congressional Recognition & Award (November 2025) — Rep. James Walkinshaw (VA-11), for excellence in biomedical engineering; entered into the Congressional Record
• Dr. Edgar A. Love Award for Excellence in Biomedical Engineering (November 2025) — Howard University, Omega Psi Phi / Omega STEM Foundation
• United Nations Delegate & Speaker (June 2025) — 18th Convention on the Rights of Persons with Disabilities, representing Jaipur Foot USA
• Startups @ Spring — 1st nationally of 700+ (2026); 2nd nationally of 700+ (2025)
• Valedictorian, BASIS Independent McLean, Class of 2026

Python · PyTorch · NumPy/SciPy · C/C++ · Arduino/ESP32 · MATLAB · R · SLURM and GPU/HPC clusters · TypeScript/Next.js · LaTeX · Git · Linux

• Email — aaryasinh.vaghela@gmail.com
• Website — avaghela.com
• LinkedIn — aaryasinh-vaghela
• Google Scholar — profile
• ORCID — 0009-0002-7942-8693