Loading...
Loading...
I design the platform decisions behind AI-ready infrastructure: which trade-offs to make, which abstractions to build, and how to make GPU clusters, inference gateways, and data platforms work as a coherent system.
My side projects are not disconnected experiments. Kortex governs inference routing and tracks costs per request. The AI FinOps Platform turns that cost data into actionable budgets and anomaly alerts. The MLOps Platform provides the full-stack reference for how model serving, security, and observability fit together across clouds. Together, they form a coherent vision for what AI-ready platform engineering looks like.
Multi-provider routing, failover, and per-request cost tracking as a Kubernetes operator.
GPU utilization monitoring, budget forecasting, and ML-based anomaly detection for AI spend.
Multi-cloud model deployment with defense-in-depth security and GitOps-driven infrastructure.
Cross-provider GPU price aggregation for finding the cheapest spot instances.
Docker Compose for AI agents with GitOps-native manifest generation.
11 professional certifications across AWS, Azure, and Kubernetes ecosystems