How To Sample Your Way Into Practical AI Inference
Talk on sampling-based approaches for practical AI inference at IC3 Blockchain Camp 2026.
Professional Summary
Pranay Anchuri is a Senior Research Scientist at Offchain Labs. His research spans
blockchain protocols, verifiable computation, and machine learning applied to
decentralized systems. He has published at top venues including KDD, JMLR, and ICDM,
and is an inventor on seven US patents. He holds a PhD in Computer
Science from Rensselaer Polytechnic Institute.
Education
PhD in Computer Science
2010
2015
Rensselaer Polytechnic Institute
B.Tech in Computer Science
2006
2010
International Institute of Information Technology, Hyderabad
Interests
Blockchain protocols
Verifiable computation and zero-knowledge proofs
Machine learning for decentralized systems
Graph mining and network analysis
Research
My research sits at the intersection of blockchain protocols, verifiable computation, and machine learning. At Offchain Labs, I work on making decentralized systems more secure and efficient. Previously at Princeton CITP, I applied machine learning to questions in public policy and social science.
I am always interested in collaborating — reach out via email or Twitter.
Recent Publications
(2026).
Towards Verifiable AI with Lightweight Cryptographic Proofs of Inference.
In SaTML 2026.
(2026).
RAmmStein: Regime Adaptation in Mean-reverting Markets with Stein Thresholds — Optimal Impulse Control in Concentrated AMMs.
arXiv 2602.19419.
(2022).
A Multi-Objective Active Learning Platform and Web App for Reaction Optimization.
JACS 2022.
(2017).
Structural Event Detection from Log Messages.
In KDD 2017.
(2014).
Hotspot Detection in Service-Oriented Architectures.
In CIKM 2014.
In the Media
- Crypto has limited utility in solving AI’s trust and payment issues, IC3 researchers say — The Block, 2026
- IC3 on AIC3 — @initc3org on X, 2026
Talks
Towards Verifiable AI with Lightweight Cryptographic Proofs of Inference
Talk on lightweight cryptographic proofs for verifiable AI inference at SaTML 2026.
Verifiable AI Inference
Talk on verifiable AI inference at IC3 Winter Retreat 2026.
Introduction to Blockchain, Ethereum and Smart Contracts
A two-session course on blockchain fundamentals, the Ethereum ecosystem, and smart contract development. Taught three times as part of Princeton's Wintersession program (2020, …
Side Projects
Speaker Randomizer
Web app for randomizing speaker order using Drand's distributed randomness beacons. Generates verifiably fair orderings that no single party can manipulate, with cryptographic …
•
1 min read
WWPotholes
Community pothole reporting platform for West Windsor, NJ. Residents pin locations, rate severity, and upvote the worst spots. Features a heat map, leaderboard, and automated …
•
1 min read