Hardhik Mohanty
Ph.D. Student, USC Machine Learning & Quantitative Research for DeFi
Research Interests
-
Machine Learning for DeFi Markets
Risk, fraud, and user-behavior models learned from on-chain transaction data.
-
Quantitative Modeling of Crypto Markets
Market microstructure, stablecoin dynamics, and liquidity under stress.
-
Game Theory & Mean-Field Dynamics
Agent-based equilibria for arbitrage, market making, and peg restoration.
-
Crypto Volatility Forecasting
Macro signals from prediction markets that sharpen realized-volatility forecasts.
News
-
New preprint on crypto volatility and prediction markets
“Do Prediction Markets Forecast Cryptocurrency Volatility? Evidence from Kalshi Macro Contracts.”
arXiv -
Paper accepted to IEEE ICBC 2026
“Who Restores the Peg? A Mean-Field Game Approach to Model Stablecoin Market Dynamics.”
arXiv -
Joined Coinbase as a part-time MLE
Building foundation risk models for users in DeFi markets using machine learning.
-
Teaching Assistant — EE 503 Probability
TA for the PhD-level probability course at USC, under Prof. Bart Kosko.
-
Summer ML Engineer intern at Coinbase
Worked on Coresets and Graph Machine Learning with Dr. Indrayana Rustandi.
-
Preprint — Survey on Zero-Knowledge Proofs
“A Survey on the Applications of Zero-Knowledge Proofs.” With R. Lavin, X. Liu, L. Norman, G. Zaarour, and B. Krishnamachari.
arXiv -
Paper in Frontiers in Blockchain
“Modeling and Analysis of Crypto-Backed Over-Collateralized Stable Derivatives in DeFi.” With Z. Feng and B. Krishnamachari.
Frontiers