You’re running several SaaS and eCommerce platforms, and the work that keeps them healthy is unglamorous: SQL that’s actually correct, product kits and relationships built right, and bad data caught before a customer ever sees it. That takes careful query work plus the judgment to sense when a number looks wrong.
That’s the part I’m good at. At DOST-ASTI I did database query optimization through in-depth inspection and root-cause analysis, integrated the core RDS database with caching and a message broker, and kept containerized services running in production. The job wasn’t writing code once — it was keeping a live system accurate and fast.
I also build with Claude Code every day, but I direct it. On Aqualytix (a production PostgreSQL monitoring app) I built a pipeline that validates incoming readings against per-station thresholds and flags anomalies — the same “check the data, surface what’s off, recommend a fix” loop you’re describing for product data.
My raw coding background is what lets me use Claude Code safely on something as unforgiving as data accuracy — I catch it when it drifts. I’m in Manila and overlap US Eastern hours daily, and I’m looking for exactly this kind of long-term engagement.
Want to point me at one messy dataset so I can show you how I’d validate it?