It’s been a full year since I quit my well-paying job to venture out on my own.
I co-founded Vaatun โ a company building software for a market most of tech ignores: Indian insurance distributors.
A year in, one takeaway from my Microsoft days has stayed with me more than any other โ Satya Nadella often reminded us that we spend more time at work than we do with our families or anywhere else, so we’d better make the most of it. Working with talented, driven people is a privilege. And the ability to choose who I work with is the single greatest privilege โ and the single greatest leverage โ I have as a co-founder. I don’t intend to waste it.
Which brings me to why I’m writing this: we’re hiring engineers across levels โ from entry-level interns to senior engineers.
Increasingly, we build interfaces for machines, not just humans. And underneath it all sits our core bet: ๐๐ต๐ฒ ๐ฏ๐ฒ๐๐ ๐๐ฎ๐ ๐๐ผ๐ณ๐๐๐ฎ๐ฟ๐ฒ ๐ถ๐ ๐๐ต๐ฒ ๐ผ๐ป๐ฒ ๐ฝ๐ฟ๐ผ๐ณ๐ฒ๐๐๐ถ๐ผ๐ป๐ฎ๐น๐ ๐ต๐ฎ๐๐ฒ ๐๐ผ ๐๐ฝ๐ฒ๐ป๐ฑ ๐๐ต๐ฒ ๐น๐ฒ๐ฎ๐๐ ๐๐ถ๐บ๐ฒ ๐ถ๐ป ๐๐ผ ๐ด๐ฒ๐ ๐๐ต๐ฒ๐ถ๐ฟ ๐๐ผ๐ฟ๐ธ ๐ฑ๐ผ๐ป๐ฒ. Time-in-app is a cost to our users, not a metric to celebrate โ a deliberate departure from the engagement and DAU numbers most of software worships.
So what do we actually screen for?
– ๐๐ผ๐บ๐ฎ๐ถ๐ป ๐ฐ๐๐ฟ๐ถ๐ผ๐๐ถ๐๐. Insurance is unfashionable. Our moat is domain depth, and we think about insurance at every step โ as CTO, even the TV shows I watch and the video games I play have somehow become insurance-centric ๐ If broker workflows and reconciliation bore you, the cutting edge tech or agentic infrastructure won’t save the marriage. We take this seriously enough that every hire โ regardless of role โ is expected to clear a few insurance certifications after a reasonable period at the company (most people pass within the first 12 months, and tend to be super proud about it). The interview tell: do you ask about our stack, or about our customers and their problems?
– ๐๐ผ๐๐-๐ฝ๐ฒ๐ฟ-๐ฑ๐ฒ๐ฐ๐ถ๐๐ถ๐ผ๐ป ๐ถ๐ป๐๐๐ถ๐ป๐ฐ๐. Building lean changes the engineering job. Every engineer here internalizes token economics and infra tradeoffs, because there is no “someone else’s budget.”
– ๐๐ผ๐บ๐ณ๐ผ๐ฟ๐ ๐๐ถ๐๐ต ๐ฟ๐ฒ๐๐ฒ๐ฟ๐๐ถ๐ฏ๐น๐ฒ ๐ฏ๐ฒ๐๐. Some of what we build gets killed. We want people who can build well anyway, without over-engineering or getting attached.
When AI absorbs implementation, ๐๐ฎ๐๐๐ฒ ๐ฏ๐ฒ๐ฐ๐ผ๐บ๐ฒ๐ ๐๐ต๐ฒ ๐๐ฐ๐ฎ๐ฟ๐ฐ๐ฒ๐๐ ๐ถ๐ป๐ฝ๐๐. A small team with excellent judgment plus serious agentic leverage beats a larger conventional team. Which means we can afford to be slower and pickier per hire โ and we intend to be.
We’ve even redesigned the hiring loop around all of this. Less LeetCode. More “decompose this messy real-world workflow into specifiable tasks.” And at the heart of it, an assignment that works both ways: we get to gauge your technical capability, and you get to gauge whether insurance as a domain actually appeals to you.
One pattern we didn’t expect: people who grew up around insurance or wealth management โ a parent in the industry, dinner-table conversations about premiums and commissions โ relate to these problems almost instantly. If that’s you, you already understand why this matters. I’d especially love to hear from you.
If this sounds like the kind of place you want to build at โ working on AI-native software for a market everyone else ignores โ my DMs are open. Whether you’re an intern starting out or a senior engineer looking for your next challenge, I’d especially love to hear from you.
Originally published on LinkedIn in July 2026.