Hi, I’m Harsha.

I’m a Senior Engineer at Cisco Systems, where I work on cloud infrastructure security. I build tools that measure what actually matters in production Kubernetes environments – not just what vulnerabilities exist, but which ones are loaded, which ones are reachable, and what happens when something gets compromised.

What I’m Working On

Plexar – An open-source security posture assessment tool for Kubernetes. It combines runtime process introspection, blast radius analysis, compliance mapping, and delegation chain security for AI agent workloads. Apache 2.0 licensed.

Ship & Sleep – A weekly newsletter about the gap between security research and production infrastructure. Kubernetes security, AI agent infrastructure, and what breaks when theory meets reality.

Research Interests

  • Runtime vulnerability analysis – determining which CVEs are actually exploitable in a specific deployment, not just theoretically severe
  • AI agent security on Kubernetes – delegation chains, blast radius in multi-agent systems, the confused deputy problem in MCP architectures
  • Security posture assessment – composite risk scoring that accounts for network topology, RBAC, workload classification, and runtime state
  • The deployment gap – why security enforcement mechanisms from research rarely reach production infrastructure

Background

Before Kubernetes security, I worked on network infrastructure and cloud automation at Cisco. I’ve been in production infrastructure engineering for 7+ years, primarily focused on the intersection of security and operations in enterprise environments.

I hold a Master’s degree in Computer Science.

Contact