0
Faster decision cycles
6+ years turning finance, operations, procurement, and supply-chain data into dashboards, automated reporting, and decision-support systems teams actually rely on.
Most teams aren't short on data. They're short on clarity. I build the systems that close that gap.
I take raw operational data and turn it into reporting people trust and act on — defining KPIs with stakeholders, modeling the data underneath, and making sure every number can be traced back to its source.
As Data Analytics Lead at Ayra Builders & Developers, I built forecasting models across $600K+ operating portfolios that identified $90K in annual savings and cut executive decision time by 40%. As co-founder of Buildbean, a B2B analytics startup, I built reporting that cut errors 60% and supported 20%+ margin improvement. I got my start as an analyst at Lokal (Y Combinator W19).
I'm finishing my M.S. in Business Analytics at UMass Amherst (Isenberg). The problems I like most sit where messy real-world data meets decisions that actually matter.
Core, everyday tools in green. Everything here is something I've used on real work or shipped projects.
Three live analytics platforms. Each starts from a real business problem and ends in a decision. Live links and code below.
A self-service platform that estimated $5.58M in potential revenue leakage across 4,121 flagged transactions — pinpointing excessive discounting as the largest driver. Built on a dbt + DuckDB warehouse with star-schema marts, Airflow orchestration, 39 automated data-quality tests, and exact-dollar reconciliation, so every number on the dashboard traces back to a tested source.
An end-to-end S&OP decision-support platform on a simulated 120-SKU supply chain (12 regions, 8 suppliers, 2 plants, 3 distribution centers, ~400K synthetic records). Pairs multi-model P10/P50/P90 demand forecasts with an OR-Tools optimization model — recommending overtime, expediting, and inventory transfers under capacity constraints, and improving modeled fill rate from 69.7% to over 99%.
A marketing analytics platform on 171K+ campaign records with real statistical rigor — 14 A/B experiments using two-proportion z-tests, Wilson intervals, and power analysis. Compares 5 attribution models alongside funnel, segmentation, and cohort analysis, producing executive summaries and channel-level budget-allocation recommendations across 25+ governed KPIs.
Open to full-time roles in Data Analytics, Business Intelligence, and Supply Chain / Operations Analytics. If you're building something and want someone who ships real, trustworthy work — let's talk.