About

I am a builder who likes end-to-end ownership: from the cron job or BigQuery pull to the operator-facing UI or browser extension. These projects sit at the intersection of messy real-world data and software that has to stay reliable week after week.

Skills I have leaned on—often alongside AI

• Node & Express for lightweight APIs, dashboards, and glue services that sit between spreadsheets, databases, and external APIs.
• Playwright for flows where the browser is the source of truth: retail configurators, promo paths, and internal tools that were never meant to ship a public API.
• BigQuery & SQL for merchant-rate and feed analysis—drafting queries, shaping aggregates, and iterating on audits without losing sight of cost and correctness.
• Google Sheets & cron so stakeholders get familiar surfaces and scheduled outputs without standing up heavy BI for every problem.
• PostgreSQL & SQLite when state needs to live closer to the app: dashboards with history, or local run logs for automation suites.
• Browser extensions to put audits and helpers directly where people already work, instead of asking them to context-switch.

How AI makes my day-to-day easier

I use AI as an accelerator, not a substitute for judgment. It helps me spin up scaffolding faster (routes, scripts, test harnesses), explore unfamiliar libraries with concrete examples, and refactor with a second set of eyes on edge cases. It is especially useful for boilerplate, repetitive refactors, and turning a rough idea into something I can run and then harden.

Day to day, that means less time stuck on ceremony and more time on the parts that matter: domain rules, failure modes, and how a tool will behave when a feed is late or a selector changes. I still own architecture, security-sensitive paths, and what ships to production—but I get to those decisions faster, with clearer options on the table.