Principal Engineer @ Microsoft Core AI — building the infrastructure that runs AI agents securely at scale.

I work at the intersection of three of the hardest problems in AI infrastructure today: agentic AI · secure agent sandboxes · reliability at scale. Here is my LinkedIn profile.

• 🧱 Azure Container App Sandboxes (Azure Dev Compute) — secure, snapshot-capable microVMs for running AI agents and model-generated code. I own the developer-facing stack: control-plane APIs, multi-language SDKs (C#, Go, Python, TypeScript), and the service resiliency that keeps a global control plane dependable.
• 🤖 Ex — Azure SRE Agent — led engineering for an AI-first, zero-ops agentic platform: autonomous monitoring, root-cause analysis, and guarded, human-in-the-loop mitigation, with evals & safety as first-class gates.

The through-line: from the agent brain (SRE Agent) to the secure platform agents run on (ADC).

AI-native — Claude & Copilot agents wired into my dev loop for full-loop automation (generate → test → evaluate → ship), with prompts, tools, and chains treated as versioned, testable, observable software, and evals as the quality gate.

Agentic AI · Secure Agent Sandboxes · Agent Runtimes · Control-plane APIs · SDKs (C#/Go/Python/TypeScript) · Distributed Systems · Service Reliability (SRE) · Kubernetes / AKS · microVMs · Model Context Protocol (MCP) · LLM Evals · AIOps / Zero-Ops

• 🧪 Azure Container App Sandboxes — https://sandboxes.azure.com/ (secure microVM agent compute)
• 🚑 Azure SRE Agent — AI-first, zero-ops agentic SRE for Azure - https://sre.azure.com

Building the agents — and the platform that makes them run securely at scale.