Interactive prototypes that demonstrate enterprise AI governance and agentic system architectures — built to be tested, not just read about.
Federal agencies are deploying AI agents to process procurement payments autonomously. But existing fraud controls — manual audits, mutable logs, retrospective reviews — were designed for humans, not machines executing hundreds of transactions per minute. The result: $236 billion in improper payments annually, with AI acceleration making the problem worse.
Cryptographic Intent Verification (CIV) ensures every AI agent proves its identity cryptographically before initiating any payment — private keys stored in hardware security modules, intent payloads signed and logged immutably.
On-Chain Policy Enforcement (OCPE) evaluates 10 smart-contract rules in real time — from SAM.gov vendor registration to Antideficiency Act budget checks — blocking fraudulent transactions before funds move, not months later during an audit.
Connect with me to explore the CIV + OCPE architecture in detail — including the working prototype.
Let's Connect →When the Medicaid continuous enrollment provision ended in April 2023, states began re-verifying 94 million enrollees simultaneously. Over 25 million Americans lost coverage — roughly 70% for procedural reasons like missed paperwork, not because they were ineligible. The crisis revealed that most Managed Care Organizations were running renewals with spreadsheets, manual call lists, and heroic individual effort. RenewIQ replaces that with a system.
Risk-scored renewal intelligence — ingests state eligibility data and MCO operational data, scores every member across 16 risk signals (deadline proximity, missing documents, contact validity, SDOH factors), and auto-assigns the highest-risk cases to the right outreach staff by language and capacity.
Full outreach orchestration — phone, SMS, email, and physical letter dispatch from a single workspace, with personalized templates, outcome logging, and audit trails that prove compliance to the state. Every action is traceable from score to outreach to resolution.
Americans lost Medicaid coverage
Lost for procedural reasons
Risk signals scored per member
Total addressable market
State eligibility feeds + MCO operational data → 16-signal Risk Engine → member score 0-100 → tier assignment (High / Medium / Low)
Auto-assign by language + capacity → 4-channel outreach (Phone, SMS, Email, Letter) with personalized templates → structured outcome logging
RBAC with enterprise SSO (SAML/Azure AD) → immutable audit trail → 9 MCO integrations (CRM, call center, DW, state systems) → compliance reporting
Additional working prototypes across agentic system governance, multi-agent orchestration, and enterprise AI architecture will be added here as they're built.