Voyages By Dave: A suite of AI-powered travel planning tools that helps savvy travelers research more comprehensively than a Google or ChatGPT search. It brings together real-time weather, travel advisories, local festivals and events, destination context, and vendor-specific knowledge into one intelligent planning experience.
MCP & LangGraph Workflows: Build agentic pipelines that pull live data from multiple sources, so LLM answers are grounded in real information instead of invented context.
Turn scattered documents into attributable answers. This self-hosted RAG stack scrapes, chunks, embeds, and reranks content so you can query a private knowledge base and get curated responses with numbered, verified sources.
Live Demo: Query the demo knowledge base in plain language and get curated, attributed answers with numbered sources.
Open source repo: FastAPI backend, admin UI, query UI, and Docker Compose setup, including a security and threat model discussion.
An evidence-driven checker that vets whether a company is Canadian. An agentic LangGraph workflow finds the official site, scrapes corporate pages, and a local Ollama model classifies strictly from collected evidence, with results streaming to a React UI in real time.
Live Demo: Enter a company name or URL and watch the evidence stream in.
Open Source Repo: Built with FastAPI, LangGraph, React, SearXNG, Firecrawl, and Ollama. Features deterministic validation — every cited quote is verified against scraped content, so no evidence, no verdict.
Operational Leadership: Directed compliance and risk management in clinical healthcare and commercial diving—environments where a missed check has immediate consequences. That zero-fail discipline now informs how I build AI governance that works under pressure.
AI Security Translation: Turn operational discipline into practical AI guardrails—governance that keeps systems secure and useful without slowing delivery or hiding risk.
Open Source Repo: A Python audit tool that traces Tableau calculated fields back to database columns, so engineers can refactor schemas without breaking dependent dashboards and reports.
Microsoft: Restructured the partner program around a depth-and-breadth model, concentrating engineering and field investment on strategic partners while maintaining broad ecosystem coverage. Contributed $5 million in annual incremental revenue, and worked across engineering, licensing, and legal teams to craft custom OEM agreements for embedded telecommunications products.
Siemens: Equipped global sales teams to drive incremental Microsoft software revenue, and reduced R&D costs by 66% by consolidating OS/2, UNIX, and Windows telephony platforms into a single platform while preserving required customer functionality.