About

I'm a systems architect with over a decade of experience designing, building, and operating large-scale infrastructure in enterprise environments. My work sits where security and automation meet: zero-trust networking, identity, and the platform automation that makes both hold up at scale.

Platform and infrastructure automation

Much of my career has been spent on platforms at enterprise scale: application platforms, container orchestration, software-defined networking, configuration management, and public cloud. The recurring problem is the same one: how to turn an architecture standard into something that is built the same way every time, by automation rather than by runbook, and stays that way. I've been recognized internally for contributions to platform modernization, network isolation, and infrastructure design.

Identity and security

My background in security started with enterprise identity and access management: federation with SAML, OpenID Connect, and OAuth 2.0; PKI and certificate lifecycle; secrets management with HashiCorp Vault; and hardware security modules for key protection. That experience shapes how I think about networks today. A connection should be allowed because of who the workload is, not where it happens to sit.

Payments infrastructure

I've also worked on payments infrastructure: high-availability database replication, HSM-backed cryptography, and the operational discipline that regulated, always-on systems demand. It taught me to treat change safety as a design requirement rather than a process step.

What I'm working on

These threads come together in Innerwall, an open-source microsegmentation platform I'm building. It applies identity-based policy to east-west traffic and rolls it out in three stages: visibility, simulation, then enforcement. I've written a seven-part first-principles series on microsegmentation drawn from its design decisions.

In my day-to-day role, I work on enterprise-scale systems and platform initiatives at Mastercard, contributing to the design and evolution of large, distributed environments.

Writing

Earlier posts here cover AI infrastructure: gateways, cost control, and governance for LLM traffic. It's the same problem from a different angle. Once a new kind of workload reaches production, it needs a control point where policy, identity, and observability are enforced. My writing draws on general systems engineering principles and publicly available information.

Contact

The best way to reach me is via email at hello@avinashpapineni.com.

You can also find me on LinkedIn.