Software Engineer in test

I build the APIs and data infrastructure that institutional clients depend on for real-time crypto market data. As a Software Engineer at Amberdata, I work across the full stack of a high-throughput market data platform — designing Go microservices, optimizing Trino/Iceberg query pipelines, and owning the CI/CD infrastructure that ships it all reliably. One of my most impactful contributions was investigating a Trino performance bottleneck on our order book snapshot endpoint — running a 48-scenario matrix across backends, identifying 5 compounding root causes, and delivering a 230× reduction in API error rates, from 23% down to 0.1%. My engineering work spans both sides of quality and delivery. I design service layers and API pagination schemes in Go, architect data quality frameworks that unify notebook-based validation with production pipeline expectations, and own monorepo release pipelines from dependency resolution through Kubernetes deploy verification. What sets me apart is the ability to move fluidly between layers — one day tuning Trino worker threads and Kubernetes node affinity, the next designing cursor compression schemes for REST APIs or publishing shared Python packages to an internal registry. I also leverage agentic AI workflows heavily in my day-to-day, using parallel Claude Code sub-agents for root-cause investigation and performance analysis — turning what used to be multi-engineer efforts into single-owner deliverables. Core stack: Go · Python · Trino · Apache Iceberg · Kafka · Spark · AWS · Kubernetes · GitHub Actions Open to backend engineering and platform engineering roles where reliability, performance, and scale all matter.