Work

Production systems that changed throughput, latency, or reliability.

I build production systems for operations-heavy teams where throughput, realtime feedback, and delivery reliability have to hold up under live usage.

Shipped systems with operating leverage

Three shipped systems where the architecture had to remove delay, backlog, or failure-prone manual work.

Ops Throughput

Automated Fundraiser Acknowledgment at Scale

OutcomeRoughly 12x per-operator throughput with 72 concurrent workers, clearing a persistent backlog.

Replaced a manual outsourced thank-you workflow with browser automation that could keep up with high-volume fundraiser traffic.

Realtime Donations

Realtime Fundraiser Creation + Donation Signal Pipeline

OutcomeEliminated a 48-hour reporting lag and surfaced donation context in realtime for operations teams.

Built a realtime fundraiser creation and donation pipeline so teams could act on live signals instead of waiting for delayed files.

Publishing Pipeline

Custom Substack Publishing Pipeline in Rust

OutcomeDelivered a deterministic pipeline from Directus webhooks through queued workers to publish actions.

Built a Rust monorepo that connected CMS events, queued workers, and publishing automation for Substack-oriented content operations.

Open-source design work

I keep the crate deep dives on a separate page so this page stays focused on shipped systems. If you want the API and code-level view of how I model invariants and tooling, start there.

Explore open-source work