Blog
Field notes on AI platform engineering, sovereign infrastructure, and messaging at scale.
6 min read
How I cut enterprise LLM costs by 40% with Kubernetes and LiteLLM gateways
Developers calling frontier models directly is how AI budgets quietly explode. Here's the gateway pattern I deploy to put spend back under control.
LiteLLMKubernetesAI FinOps
7 min read
Building sovereign AI: why hybrid Kubernetes is the future of enterprise data
Banks, healthcare, and governments can't send PII to public APIs. Here's the air-gapped, hybrid architecture that keeps data inside the perimeter.
Sovereign AIComplianceKubernetes
5 min read
Surviving Meta's rate limits: engineering WhatsApp for 500+ messages/second
Anyone can call the WhatsApp Cloud API. Surviving high-throughput bursts without dropping receipts is the real engineering.
WhatsAppKafkaDistributed Systems