A Go pricing service's goroutine count grew by thousands per hour. A helper that raced a slow lookup against a timeout left the lookup goroutine blocked forever on an unbuffered channel nobody would read. How we found it with goroutine profiles and fixed it with context and goleak.
A Go ingestion service that forwarded events to a partner API was OOM-killed every time traffic spiked. Each event started a goroutine that waited on the slow partner, so goroutines and their buffers grew without limit. How we found it with pprof and fixed it with a bounded worker pool.
We designed a multi-tenant payment platform for debt-collection agencies: card and ACH, one-time and recurring plans, settlement offers, disputes and reconciliation. Here is how we kept PCI scope small, why the ledger is double-entry and immutable, and how retries stay safe.
A reminder job sent 380 emails to one customer and none to the rest. The code was correct under Go 1.22 loop semantics, but the module's go.mod still said go 1.20, so every goroutine shared one loop variable. Here is the full trace and fix.