I'm Aman. I work on infrastructure at TCS in Bengaluru — pipelines, deployments, databases, and the machinery that keeps the product running.

Like a lot of people, I started as a developer

My degree had an iOS specialisation, so that's where I began. My first project at TCS was a SwiftUI iPad app for a client. I built the whole thing and it shipped in about two months.

Then the feature requests dried up and someone handed me the AWS side of the same engagement, mostly because there was nothing else for me to do. That's the part that stuck. I asked to move.

What happened after that

I joined the Platform Solutions unit expecting to build a mobile client for TCS Enterprise Manager. Budget said otherwise, and I landed in a DevOps role right as the unit decided to fix how it shipped software.

What we walked into: 75+ microservices sitting in SVN, deployments every Thursday, no review gate, and a revert that took an hour of careful manual work. Two of us were told to sort it out.

We spent longer than anyone expected just writing down how delivery should work before touching Jenkins. Which branches mean what. What a merge has to pass. Who's allowed to deploy to prod. It felt slow at the time. It was the right call, because every pipeline after that was just implementing the document.

The revert pipeline is the bit I'm proudest of. Getting it under thirty seconds is what actually changed how people behaved — once undoing a release stopped being frightening, nobody wanted to wait for Thursday any more.

Since then it's mostly been databases and access. Superset in front of DEV, UAT and prod so nobody needs raw credentials. Bytebase handing out logins that expire on their own. An artifact registry decision I'd happily defend in an interview. A lot of Terraform.

Books

I read constantly, and it's the fastest way to get a sense of how I think. Mostly non-fiction and systems writing, with science fiction when I want to stop thinking about work entirely. My Goodreads is public — what I'm reading now, what I've finished, and what I gave up on. If you're hiring, it'll tell you more than a skills list will. If you just want a recommendation, I always have one.

The homelab

I self-host more than is strictly reasonable. Identity, reverse proxy, search, monitoring, media, all sitting behind Cloudflare tunnels with the config in version control. Most of it exists because I wanted to understand something properly, and breaking it at home costs a lot less than breaking it at work.

I'm writing Flutter for the product's mobile client at the moment, so I haven't entirely left app development behind.

---

Come say hello

I'm open to DevOps, SRE, platform and cloud architecture roles. I'm also happy to talk about self-hosting setups, or argue about books.