Likhith Madhav Vankadari

My Story

I'm currently the person responsible for keeping things running, and keeping costs under control, for a fast-growing Web3 gaming platform with over 200,000 players worldwide.

In a system where traffic spikes without warning and downtime is immediately visible to users, I focus on building infrastructure that doesn't just work, but holds up under pressure. I design and operate Kubernetes-based systems, run GitOps-driven deployments, and constantly look for inefficiencies. Recently, that meant cutting our AWS bill nearly in half without sacrificing performance.

But the real work often begins when things go wrong. I've handled live security incidents, including rotating compromised API keys in production and rebuilding parts of our email infrastructure on the fly. I've tightened IAM in a high-risk environment and built observability systems so we can detect issues before they escalate. These experiences shaped how I think: systems aren't defined by how they behave on a good day, but by how they respond when everything is under stress.

Before this, I worked in a completely different environment: an air-gapped defense setup where nothing external could be trusted or accessed. No internet, no quick fixes. I built internal toolchains from scratch, hosted dependencies locally, and ran fully isolated CI/CD pipelines. From installing VMware ESXi on bare metal to standardizing deployments, I learned how to build reliable systems when the easy path simply doesn't exist. That work reduced deployment cycles from hours to minutes.

Over the last four years, the tools have changed, Jenkins to Terraform, scripts to Helm, but my approach hasn't: build for failure, understand how systems break, and automate the recovery.

Right now, I'm pushing deeper into Site Reliability Engineering, with a focus on MLOps and AIOps. I'm interested in systems that don't just react, but anticipate: infrastructure that can detect anomalies, predict failures, and heal itself before anyone notices.

At the core, I care about one thing: building infrastructure that lets teams move fast, stays resilient under pressure, and quietly does its job so well that nobody has to think about it.