At Aurva I work across the full data security stack — DSPM, DAM, CIEM and AISPM — used by multiple enterprises in the APAC and EMEA regions. I've written custom parsers for every major cloud IAM service based on their grammar, implemented finite state machines for wire protocol parsing and dealt with extracting prompts from encrypted API calls.

I currently lead AI security development at Aurva and I'm in the phase I find most interesting: building entire product verticals from scratch — identifying the problem, making MVPs, iterating fast, and talking to customers. The full loop, not just the code.

The question I keep coming back to: what actually happens when AI agents meet production environments? The attack surface is larger than most teams realise, the tooling is still catching up, and the people who should be worried often aren't yet. Someone needs to work on that problem seriously.

I started with Linux at 16 and never really stopped. Software, hardware, mechanical systems — whatever the problem demanded. The question I keep asking is "what if we just built this ourselves?" — sometimes the answer is no, usually it's more interesting than expected. Engineering isn't just what I do, it's the default mode. I owe my grandfather for that.

Outside work: Hacker News addict, relentless tracker of product launches, funding rounds, and M&A deals. I read a crazy amount of books - already did 8 in 3 months of 2026 - host quizzes and argue in debates because sharing curiousity and viewpoints is a wonderful thing. I also have a genuine interest in advertising, history, and talking to people — which makes me the king of random, apparently.