Piyush Kumar

Piyush Kumar is a Finance and AI Specialist whose work bridges the gap between traditional financial engineering, healthcare economics, and decentralized technology. He holds a Dual Degree in Computer Science (B.Tech Hons.) and an MBA in Finance from Amity University, Noida.

Currently an independent researcher, Piyush specializes in financial market microstructure, mortgage default risk, value-based healthcare optimization, and the intersection of Behavioral Economics and Decentralized Finance.

His scholarship includes an active SSRN Working Paper Series featuring research on Hardware-Enclave Biometric Salt Generator and Risk-Adaptive Deterministic Cross-Chain Identity Engine (WIPO Patent Pending: PCT/IB2026/058453), multi-stakeholder equilibrium models for precision capitation, and diagnosis-contingent payment models.

His work spans on engineering high-precision solutions for the global financial landscape, specifically through Hormone-Responsive Logic engine, Behavioral-Microstructure Matching Engines, Risk Adaptive Stablecoin Protocols, and GNN-powered auditing pipelines. His core includes using GARCH modeling and Time-Series Analysis to forecast market dynamics, alongside investigating how formative economic experiences influence real-time institutional risk solvency.

CORE COMPETENCIES

Quantitative Finance

Household & Real Estate Finance: Structural modeling of mortgage option exercise under digital frictions, rental search indicators for default risk, and PropTech spatial dynamics.
Market Microstructure: Quantitative modeling of high-frequency market signals and DeFi protocol solvency using Behavioral Volatility Indices (BVI) to mitigate liquidity shocks; recently expanded to aligning Behavioral Experience Effects and Responsive Risk Management Framework with microstructure volatility to calibrate personalized hedge ratios
Healthcare Equilibrium & Capitation Modeling: Formulating multi-stakeholder equilibrium models for precision capitation, digital triaging, and value-based pricing systems.
Financial Volatility & Time Series: Expertise in GARCH, and Time-Series forecasting to predict commodity and asset price dynamics.
Fintech Analysis: UPI penetration modeling and digital payment ecosystem research.
Asset Correlation: Statistical analysis of Gold, Oil, and Currency price dynamics.

Artificial Intelligence

Graph Neural Networks (GNN): Advanced application of TensorFlow-GNN to transform compiled bytecode into directed graphs for real-time smart contract vulnerability detection.
Deep Learning: Expertise in FaceNet and Neural Networks for Decentralized Biometric Identity (DID).
Predictive Analytics: Advanced experience in sentiment analysis (DistilBERT) and classification.
Big Data in Healthcare: Foundational research in medical AI and diagnostic modeling.

Data Engineering & BI

Visualization: Expert-level Tableau developer (Keynote Speaker at NTU Singapore)
Programming: Robust technical toolkit including Rust, Python (NumPy, Pandas, Scikit-Learn), and R for blockchain and financial engineering.
Business Intelligence: Strategy development for revenue projections and operational efficiency.

Current Research & Development

Biometric DID & Cross-Chain indentity (Patent Pending): Invented a hardware-enclave biometric salt generator and risk-adaptive cross-chain identity framework (PCT/IB2026/058453).
Smart Contract Auditor: Created an AI-Driven Smart Contract Risk Engine integrated with a Solana-based Registry to identify institutional vulnerabilities.
Stablecoin Protocol: Designed a Behavioral Volatility Index (BVI) and a dynamic fee-adjustment smart contract in Rust to preserve protocol solvency during extreme market de-pegging events.
Behavioral-Microstructure Matching: Engineering a personalized financial architecture that aligns high-frequency market microstructure signals with individual "Experience Effect" biases to dynamically calibrate institutional hedge ratios.
Hormone-Responsive Logic engine: Developed an automated "Biological Circuit Breaker" that dynamically modulates portfolio leverage caps (e.g., dampening from 10x to 1x) based on real-time Biological Calm (RMSSD) levels.

LEARN MORE ABOUT WORKING PROJECTS

LEARN MORE ABOUT WORKING PAPERS

Page updated

Google Sites

Report abuse