Hi, I'm Venkatesh.

I grew up in Andhra Pradesh, with a curiosity about how things work that eventually landed me in computer science. The interest started early — not from any grand ambition, just from noticing that software was the one thing where you could build something real without needing money or materials.

I joined IIT Jodhpur for B.Tech CSE in 2019. The first year was disorienting — algorithms felt abstract, and I wasn't sure I'd ever enjoy coding outside of problem sets. Second year flipped that. We started building actual systems — small services, real data — and something clicked. The satisfaction of watching a system handle real load felt different from anything else.

By third year, I was interning at Aspiro and accidentally falling in love with backend specifically. There's something about the invisibility of it — when you do it well, no one notices, but when you mess up, the whole product breaks. That kind of responsibility appealed to me.

I joined Cars24 in 2023 as a backend engineer. The work centres around high-volume data flows — vehicle listings, pricing pipelines, event-driven workflows that need to stay consistent under load.

The interesting work has been on scaling event pipelines as inventory grew, redesigning webhook handling to deal with out-of-order delivery, and introducing smarter caching strategies at multiple layers. Most days are a mix of writing and reviewing code, debugging things that broke, and trying to predict the next thing that might break.

I'm on the SDE-2 track — which in practice means taking ownership of larger systems, helping newer engineers get unstuck faster, and being expected to make architecture decisions that hold up over time. The shift from how to why is the most interesting part of that progression.

Every engineer on LinkedIn is debating whether to use AI. I find that debate boring — the more interesting question is how you use it.

Here's roughly how I work: I reach for AI when I need to scaffold something fast, explore an unfamiliar library, or generate test cases for edge conditions I'd otherwise miss. I don't reach for it when making architecture decisions, debugging subtle race conditions, or figuring out what's worth building in the first place.

What AI does well for me: first drafts, boilerplate, documentation, test generation. What it doesn't do for me: production context, system topology awareness, deciding which tradeoffs matter.

The pattern I've found that works is this — AI handles the shape, I handle the intent. AI writes the first version, I do the critical read. The lesson I keep relearning is that AI generates working code, but working ≠ production-ready. That gap is where the senior engineer still earns their salary.

The next year or two, I want to:

• Get deeper into distributed systems design — actually intuiting CAP tradeoffs, not just memorising them.
• Ship a side project I'd be proud to show. Right now that's the content automation pipeline I've been building.
• Write more, write more clearly, write more honestly.
• Find a team where systems thinking is a default, not a fight.

Long term: I want to be a staff engineer who's known for getting hard architecture decisions right and explaining them simply. That probably takes another 4-5 years of doing the work. I'm okay with that.

I read a lot — mostly nonfiction, some philosophy, occasionally a novel I'm late to. I travel when I can; my best ideas usually come on long bus rides somewhere with bad cell service. Movies, sometimes — though I'm pickier about those than I should be.

If you're curious what I'm reading, watching, or where I went last — there's /life for that.