Projects

A deployed marketing-claim governance service: a registry that answers deterministic questions about a claim, an adversarial workflow that argues the judgment calls, and a public record an agent can fetch instead of taking vendor prose at face value.

Six-agent pipeline that classifies, drafts, generates, reviews, improves, and polishes SAFe feature specifications with a 100-point rubric scoring system.

AI-powered dining concierge for Las Vegas with Rat Pack-era personality, real restaurant discovery via Google Maps, mock booking with a real API contract, and Google Calendar deep links with insider tips.

AI-powered tool that generates structured SAFe feature specifications from natural language descriptions.

Conversational AI that replaces form-based feature request intake with an adaptive chatbot, bridging the gap between messy stakeholder input and structured specs.

RAG-powered Q&A tool grounded in team knowledge base documents.

Nine governance modules covering cost guardrails, grounding verification, content safety, bias detection, audit trails, and prompt promotion rules for production AI pipelines.

SQLite-backed evaluation infrastructure with prompt versioning, token tracking, a golden test set, and a 5-tab Streamlit dashboard for monitoring AI pipeline quality.

Multi-agent tool that models personalization investment scenarios with validated conversion economics across conservative, moderate, and aggressive projections.

Interactive reference tool for a 43-signal visitor classification system, with scoring simulator, signal explorer, and role profiles.

AI-powered advisory tool for enterprise marketing teams running buying group personalization programs, built on a synthetic B2B SaaS corpus with three interaction modes and a live website experience simulator.