Skip to content
View realgauravvyas's full-sized avatar
🏠
Working from home
🏠
Working from home

Block or report realgauravvyas

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
realgauravvyas/README.md
typing banner

Portfolio Contact DROSOMIND Hugging Face OEIS A181671 OEIS A190502

Gaurav Vyas

Student of IIT Guwahati in B.Sc. (Hons.) Data Science & Artificial Intelligence (CPI 7.5, Oct 2023 – Sep 2027), with degree research centered on deep-learning side-channel cryptanalysis (recovering a certified 128-bit AES key from physical power traces), vision-language mathematical document AI, and computational neuroscience — complemented by statistical training as a student of ISI Kolkata in the Post Graduate Diploma in Applied Statistics (Aug 2026 – Aug 2027; qualified admission test with full tuition fee waiver), alongside extending and publishing OEIS A181671 (to 10²³) and OEIS A190502 (to 2⁷²).

Interactive research portfolio indexed by Ramanujan primes running in a real in-browser sieve. Click an anchor to jump directly to the work.


🧠 AI & Data Science Research

My primary research centers on machine learning, deep neural architectures, and data science applied to physical systems, document intelligence, and neural connectomics.

Project / Repository Focus Area & Architecture Key Results & Artifacts Links
Deep-Learning AES-128 Key Recovery Side-channel cryptanalysis • Domain adaptation • ASCAD traces 16/16 key bytes recovered; $\log_2 \text{rank} \le 1$ across 3 seeds; guessing entropy 0 at 2k traces IIT Guwahati B.Sc. Degree Project
DROSOMIND Computational Neuroscience • Spiking Neural Network • LIF 166k neurons · 125M synapses; 16-wedge ring attractor compass; courtship song audio synthesis; Google Research Cell 2026 Live Simulation ↗Repo
ocr2tex Vision-Language • Document AI • GLM-OCR & Baidu OCR 0% → 82.4% compile rate (88.9% v3.1); 0.382 mean CER (25.9% relative cut); 13.9k validated pairs; GGUF models Hugging Face Models ↗Repo
chess-ai Reinforcement Learning • Policy-Value Residual Net 760k-parameter AlphaZero architecture; rigorous sample efficiency analysis and mutation testing Live Dashboard ↗Repo
drishti Probabilistic Modeling • Crisis Data Science Calibrated Bayesian belief resolution over conflicting field reports + emergency routing Repo
RouteGuard Agentic Systems • Symbolic Planning Model-based autonomous recovery; 17/17 tests, 80/80 oracle agreement; stdlib Python Live Demo ↗Repo

DROSOMIND Live Connectome Simulator


🔭 Computational Mathematics & Interactive Simulations

Turning complex mathematical and physical structures into real-time interactive systems:

🌀 Mathematical Surprises 🏛️ Millennium Prize Problems 🔶 What Survives The Sieve
250+ live interactive simulations with real-time parameter exploration All 7 problems explained with exact arithmetic & live zeta zero computation In-browser Sieve of Eratosthenes indexing 12 verified projects
Explore Visualizations ↗ Simulate Millennium Problems ↗ Visit Live Portfolio ↗

📐 Sequence Extensions in the OEIS

Official sequence extensions computed, verified with independent certificates, and published in the On-Line Encyclopedia of Integer Sequences:

Published Sequence Base / Domain Previous Record Extended & Published Record (Gaurav Vyas) Status Verification & Source
OEIS A181671 Base 10 ($10^n$) $10^{17}$ $10^{23}$ (a(18)–a(23)) Published (Aug 2026) Official b-fileGitHub Repo
OEIS A190502 Base 2 ($2^n$) $2^{56}$ $2^{72}$ (a(57)–a(72)) Published (Sep 2026) Official b-fileGitHub Repo
Definition : Rₙ is the smallest integer such that π(x) - π(x/2) ≥ n for all x ≥ Rₙ
Technique  : 128-bit Segmented Sieve of Eratosthenes + Bracketed Analytic Tail Bounds
Execution  : High-performance C++20 / Python with independent mathematical certificates

🎮 Game Development & Edge Systems

Zero-dependency offline games and field-ready applications:

Project Platform & Scope Highlights
📱 Voltfall Android Roguelite ~94 KB APK • Custom SurfaceView engine • Zero runtime dependencies • No INTERNET permission
🖥️ Neon Depths Windows Roguelite Twin-stick action • Procedural audio & geometry • 12 floors
📷 hemispheR-py Python & Android Clean-room optical canopy physics & Leaf Area Index (LAI) estimation for field ecologists
🌱 carboncampus Offline-First PWA Campus carbon footprint calculator • Shared client/server engine • IndexedDB

🛠 Toolbox

Python C++ PyTorch NumPy Kotlin TypeScript JavaScript React LaTeX Streamlit PostgreSQL GitHub Actions


🔧 How this page builds itself

The GitHub contribution snake animation is automatically drawn fresh by CI on every push to main and once daily, then published to the output branch via Platane/snk.


Followers Stars Profile views

🔶 Interactive Portfolio • ✉️ Contact


"The purpose of computing is insight, not numbers." — Richard Hamming

Popular repositories Loading

  1. ocr2tex ocr2tex Public

    LoRA fine-tunes of GLM-OCR for handwritten math to compilable LaTeX transcription

    Python 1

  2. blackout-studio blackout-studio Public

    Blackout Studio — Voltfall (Android) and Neon Depths (Windows). Zero-dependency, fully offline games.

    JavaScript 1

  3. sa_course_25 sa_course_25 Public

    This is a repository for IITG Summer Analytics Course.

    Jupyter Notebook

  4. Gemini-AI-Studio Gemini-AI-Studio Public

    Gemini 3 powered assignment evaluator

    TypeScript

  5. hemispheR-py hemispheR-py Public

    Dependency-light Python port of hemispherical fisheye canopy photo analysis — LAI, clumping index, gap fraction, no OpenCV/R required

    Kotlin

  6. mathematical-surprises mathematical-surprises Public

    Interactive mathematical & physics visualizations

    JavaScript