Governed AI workflows built with human-in-the-loop controls, validation, and auditable behavior.
01
DataReady Autopilot
Founder / Technical Program & Product Lead
Deployed to Google Cloud Run
Built and deployed a constrained Gemini repair-planning workflow with Python service logic, deterministic policy/validation controls, automated tests, source preservation, and human-controlled execution; deployed to Google Cloud Run.
System / Workflow
Constrained Gemini repair-planning workflow with Python service logic.
Controls
— Deterministic policy/validation controls
— Source preservation
Human Oversight
Human-controlled execution.
Validation
Automated tests.
Technology
— Gemini
— Python
— Google Cloud Run
02
DecisionReady
Founder / Technical Program & Product Lead
Built and deployed
Built and deployed a governed AI decision workflow with explicit NOT_READY → READY → HUMAN APPROVED states.
System / Workflow
NOT_READY → READY → HUMAN APPROVED
Controls
— Baseline/version control
— Human authorization
— Auditable decision history
— Preventing recommendations from bypassing governance
Human Oversight
Human authorization is required before a decision is approved.
GCP-based trusted-ingestion and AI workflow using Gemini and agent-oriented components, emphasizing provenance, controlled processing, traceability, and auditable system behavior.
System / Workflow
Trusted-ingestion and AI workflow using Gemini and agent-oriented components.
Controls
— Provenance
— Controlled processing
— Traceability
— Auditable system behavior
Technology
— Google Cloud (GCP)
— Gemini
— Agent-oriented components
04
HooraAI
Founder / Technical Program & Product Lead
In development — MVP / prototype
AI digital-workforce MVP/prototype translating product strategy into requirements, system architecture, phased roadmap, human-in-the-loop controls, governance, and implementation readiness.