Building AI systems that make decentralised infrastructure provably safer. MS Computer Science · Monroe University · GPA 3.8.
Peer-reviewed preprints targeting IEEE conferences. Both systems are production-deployed and publicly accessible.
Existing tools rely on rule-based static analysis that misses contextual, cross-function vulnerability patterns behind the most costly DeFi exploits. CyberTrust AI applies Claude Sonnet for semantic analysis — detecting indirect reentrancy and access control flaws that Slither systematically misses.
Real-time speech analysis and NLP classification for live sales coaching. Sub-2-second pipeline from speech to coaching insight combining Web Speech API, Next.js edge functions, Claude Sonnet inference, and PostgreSQL persistence at production scale.
Both platforms are production-deployed, publicly accessible, and actively used.
Production LLM security platform for Solidity smart contract auditing. Contextual vulnerability detection, AI-generated fix suggestions, on-chain NFT trust certificates, batch auditing, and threat simulation.
Real-time AI sales coaching SaaS. Live speech-to-text pipeline with NLP classification, objection detection, buying signal identification, and automated post-call scoring and playbooks.
A selection of research and engineering projects from my GitHub.
Core technologies and tools across security, AI, blockchain, and full-stack engineering.
Open to research collaborations, cybersecurity discussions, and speaking about LLM-based security systems.