Working projects built around business questions, data systems, AI workflows, and decision outputs.

A macOS desktop assistant that combines realtime voice, visual artifacts, local workspace context, image tools, notes, web search, and confirmation-gated computer actions.

Turns a chat interface into an operator workspace: selected files and folders become grounded context, tool results become reviewable artifacts, and risky actions stop for explicit confirmation before execution.

Production-grade AI-powered architecture and workflow builder with realtime collaboration, AI-assisted systems design, and interactive canvas orchestration.

Positions Ghost AI as an AI-native product, analytics, and systems orchestration platform for collaborative system design, product/spec generation, background AI workflows, and production debugging.

A WOO warehouse case study diagnosing why commission rows failed to match payroll employees, then fixing the workflow with deterministic SQL keys, governed overrides, and visible match confidence.

Turns an error-prone reconciliation workflow into auditable data infrastructure: normalized name keys, MAP_OVERRIDE governance, NAME_KEY_EXACT and NAME_KEY_FIRSTLAST matching, explicit UNMATCHED flags, and queryable match_method diagnostics.

A WOO warehouse case study applying control/treatment/delta discipline to financial planning scenarios, using a reconciled monthly cost baseline and an isolated vendor-spend treatment.

Productizes what-if analysis: the reconciled fact_cost_monthly_2025 baseline is the control, each analytics.scenario row is a named treatment, and delta_monthly gives a queryable effect for budget decisions.

A natural-language interface to the BudgetDB warehouse: ask a plain-English question, Claude runs read-only SQL through tool calling, and the answer is grounded in queried rows.

Extends BudgetDB from warehouse to conversational decision layer: schema-grounded tool use, read-only execution, strict query allowlisting, and visible SQL audit trails.

A quantitative product case study showing how KPI instrumentation, A/B testing, and dashboard baselines supported pricing and feature decisions for a subscription product.

Frames churn as a set of testable product levers: dashboard-tracked KPI baselines, scoped pricing and feature variants, A/B test readouts, UAT checkpoints, and roadmap decisions backed by measured outcomes.

A vendor spend and ROI analytics model: staffing-weighted budget rollup, vendor cost-per-active-employee ranking, and renewal review flags for low-adoption, high-cost tools.

Connects budget rollups, staffing assumptions, vendor utilization, and renewal review logic into an auditable operations decision model.

A payments/remittance fee-schedule and tiered revenue-projection model: NRC/MRC fee structures, transaction-tier volume assumptions, and projected monthly fee revenue by tier.

Turns remittance pricing assumptions into a repeatable revenue model with SQL rollups and Excel financial modeling artifacts.

Dashboard views for contact center performance, revenue mix, product trends, and executive operating reviews.

Makes KPI patterns easier to scan and turns analysis into a review-ready operating surface.

SQL validation checks comparing source totals to modeled fact totals by cost category, with variance and pass/fail status.

Adds confidence gates before dashboards and executive summaries are used for decisions.

A reproducible Python workflow for churn analysis: loading, target conversion, missing-value handling, encoding, baseline modeling, and reporting.

Frames predictive analysis as a repeatable business workflow instead of a one-off notebook.

An iPhone app and web landing-page concept for learning AI agent fundamentals, comparing platforms, and saving reusable skills.

Shows product thinking beyond portfolio work: mobile UX, content model, Supabase schema, launch plan, web page, and launch video assets.