[Your Name] · [Email] · [Phone] · [City, ST]
April 21, 2026
Dear Hiring Manager,
I'm applying for the Senior Backend Engineer role on your Orders Platform team. Your recent RFC on migrating from a shared Postgres primary to per-tenant schemas — which I saw linked in your engineering changelog — mirrors the isolation work I led at Shopify Plus last year, and the trade-offs you raised about cross-tenant reporting are exactly where I landed.
At Shopify Plus I owned the rewrite of our order ingestion pipeline from a synchronous Rails monolith endpoint to an event-driven Go service fronted by Kafka. We scaled peak write throughput from 8,500 to 72,000 orders per minute during Black Friday 2025 without a single dropped event, while cutting p99 write latency from 1.8s to 210ms. The interesting part wasn't the infra — it was the idempotency contract. We had to design an order-dedup key that worked across three legacy clients, two of which couldn't be changed for compliance reasons, so we built a sidecar normalizer that read their payloads and emitted a canonical form into Kafka. That sidecar has processed 4.1 billion events since launch with zero customer-reported duplicates.
Before Shopify I was backend engineer #4 at a B2B payments startup (Finix), where I wrote the first production Go service and designed the double-entry ledger that still runs today at 12M+ txns/month. I also spent a quarter embedded with our compliance team building PCI scoping boundaries — unglamorous work, but it's why I care about observability as a first-class product concern, not just an ops bolt-on. Your team's decision to run OpenTelemetry end-to-end before hitting 50 engineers is one of the reasons I want to work there specifically.
I'd welcome the chance to walk you through the idempotency design and hear where your team is in the per-tenant migration. I can share a redacted architecture doc from the Shopify rewrite or jump on a call whenever works for your schedule.
Sincerely,
[Your Name]