Dear Hiring Manager,
At Meridian Systems I built the GitHub Actions pipeline that took mean time to production from two days to under thirty minutes, with automated tests, linting, and blue-green deploys. That work taught me what Cascade Ledger needs from an engineer: a service is not done when it compiles, it is done when it ships safely and reports the truth about itself in production. I have spent five years on AWS in Go, Python, PostgreSQL, and Kubernetes, and I instrument a service before I trust it. A ledger that quietly disagrees with itself is the kind of system that habit is built for.
At Meridian Systems I designed and shipped a Go microservice handling 4,000 requests per second, cutting p99 latency on the checkout path from 520ms to 210ms with Redis caching and connection pooling. I led the migration of a monolith to event-driven services on AWS using ECS, SQS, and Lambda, which raised order-service availability from 99.9 to 99.97 percent and isolated failures to single domains. I also raised test coverage on the payments service from 61 to 89 percent with Go table tests and integration suites, which cut production incidents roughly in half over two quarters. Earlier, at Calderon Labs, I cut a nightly PostgreSQL batch job from 45 minutes to 7 by adding targeted indexes and reading the EXPLAIN output.
On the observability side, I can send a short write-up of how I added structured logging and Prometheus metrics across a request path at Calderon Labs and cut mean time to detect production issues from forty minutes to under ten, which is the same discipline a ledger under load tends to reward. I hold the AWS Certified Developer Associate and studied computer science at the University of Washington. If a thirty-minute call about how Cascade Ledger handles writes under load would be useful, I can put time on the calendar next week.