I’m focused on building "Zero-Trust" infrastructure for Sovereign AI—allowing massive-scale model training without centralized data pooling. My recent work includes validating a 500,000-node stress test with a 55.5% Byzantine Fault Tolerance boundary using Tiered Rényi Differential Privacy and zk-SNARKs.

Technical Interests:

Distributed Systems & Consensus (BFT)

Formal Verification (Coq/TLA+)

Privacy-Preserving ML (Differential Privacy, FHE)

High-performance Go & Wasm

Current Project: Sovereign-Mohawk Protocol Code: GitHub Repo

Feel free to reach out if you're interested in the intersection of formal methods and decentralized AI.