At Akamai Technologies, I wrote the request routing logic for Inference Cloud, steering AI inference traffic between edge GPU sites and core regions. The product launched on NVIDIA Blackwell GPUs and BlueField DPUs.
I also built Go services and Terraform modules for Akamai’s managed Kubernetes App Platform, and provisioned internal developer clusters through Crossplane and Argo CD. I wired OpenTelemetry and Prometheus into edge services, with that data feeding Akamai Cloud’s 2025 FedRAMP High Ready assessment.
At Amazon Web Services (AWS), I moved service tables from Oracle to DynamoDB and Aurora as part of Amazon’s internal Oracle exit. After major service incidents, I helped split a front-end fleet into cells and added reliability measures including alarms, staged deployments, and automatic rollback.

