People ask us about the Rajkot choice at every conference we go to. Usually the question is framed as a challenge — "isn't it harder to recruit there?" or "don't you miss the startup ecosystem?" The honest answer to both is yes, and we'd make the same choice again without hesitation.
Here's the full reasoning, including the parts that haven't worked out.
The burn rate difference
Office space in Rajkot costs roughly one-fifth of equivalent space in Bengaluru's tech corridors. Engineering salaries are 35–40% lower for the same skill level — not because Rajkot engineers are less skilled, but because the market isn't distorted by Google, Swiggy, and a dozen funded startups hiring simultaneously in the same few postcodes.
For a bootstrapped company, this difference is existential:
| Rajkot | Bengaluru equivalent | |
|---|---|---|
| Runway from our initial revenue | ~18 months | ~6–8 months |
| Office cost | 1x | ~5x |
| Engineering salary, same skill level | 1x | ~1.6x |
The difference between six months and eighteen months of runway is the difference between a company that survives its early stumbles and one that doesn't. We weren't being clever. We were being honest about what we could afford to build, and building there.
That constraint is inseparable from our decision not to raise venture capital — the two choices only work together.
The loyalty difference
This one surprised us. Our average engineer tenure is 3.2 years, against an industry average in Bengaluru of roughly 1.1 years.
The engineers who join us are mostly from Gujarat and Rajasthan. They have family here, roots here, plans to buy a flat and settle here. They're not building a résumé for a Bengaluru salary negotiation. They're building something in the city they've chosen.
The result is a team with institutional memory that's genuinely rare in a company our age. Engineers who've been here two years know every corner of the codebase, every past mistake, and the reason behind every architectural choice.
That knowledge doesn't show up in any metric, but it makes the team dramatically more effective than the headcount suggests — and it's a large part of why an eight-person cap is workable at all.
In Bengaluru, you hire ten engineers knowing three will leave in a year. In Rajkot, you hire seven and expect most of them to still be here in three.
The compounding effect nobody budgets for
The retention number is worth translating into what it actually buys, because "3.2 years versus 1.1" reads like a nice-to-have and isn't.
At 1.1-year tenure, a team of eight replaces roughly seven people a year. Each replacement costs a recruiting cycle, three to six months of ramp, and — the expensive part — the permanent loss of context about why things are the way they are. Teams in that state re-litigate settled decisions constantly, because nobody present remembers the original reasoning.
At 3.2 years, the same team replaces two or three people a year and everyone else is carrying four-figure hours of accumulated context.
When we decided to rebuild the entire voice stack, the people arguing about it had all personally built and operated the thing being replaced.
That argument is unavailable to a team that turns over annually, and it's the argument that saved us from rewriting the wrong layer.
The talent difference (the one nobody talks about)
The objection we hear most: "but the best engineers are in Bengaluru." This is true in aggregate and false in the way that matters.
The best engineers are in Bengaluru because that's where the highest salaries and the most prestigious logos are. The engineers who'd be equally good but haven't made that move — because they're building roots somewhere, or the metro lifestyle isn't what they want, or nobody's offered them something compelling in a Tier-2 city — exist in large numbers and are almost entirely untapped.
We've hired engineers from NIT Surat who turned down Bengaluru offers to join us. Not because we paid more, but because we offered a technical problem worth solving, the autonomy to solve it, and the prospect of staying in the city they grew up in.
That combination beats a salary bump for more people than you'd expect. The 18-month retrospective on how that hiring has actually gone is in hiring in Rajkot, 18 months later.
It also gave us something we didn't plan for: a team whose own families and neighbours are exactly the small businesses we sell to.
When we're designing for a clinic or a two-wheeler service centre in a Tier-2 or Tier-3 city, that's not market research. It's someone's uncle.
What we gave up
I want to be honest about the trade-offs, because the romanticisation of Tier-2 tech sometimes glosses over them.
Recruiting takes longer and our pipeline is thinner. There are skill sets — specific machine learning specialisations, some infrastructure experience — genuinely harder to find here.
We've had a few hires not work out because the person wanted the cultural energy of a metro startup, and Rajkot isn't that.
We're also less connected to the ecosystem. We hear about trends later. We meet fewer investors organically. Networking that happens automatically in Bengaluru costs us a flight and a hotel.
These are real costs. They're smaller than the costs of the alternative — and they haven't stopped us serving customers in 14 countries from an office here.
When Tier-2 is the wrong answer
I don't think this generalises the way the LinkedIn version of it does, so it's worth naming who shouldn't copy us.
If you need to hire fast, don't. Our model works because we hire slowly and keep people. A company that needs 40 engineers inside a year will exhaust the local pipeline in month four and end up paying metro salaries to relocate people anyway — with none of the retention benefit, because a relocated hire has no roots to keep them.
If your raise depends on ecosystem access, don't. Warm introductions, the accidental conversations at a demo day, the investor who knows three people who know you — those compound in a metro and don't exist here.
We could ignore that only because we weren't raising. If funding is the plan, proximity is part of the product.
If your talent need is genuinely specialist, be honest about it. We can hire excellent generalist engineers here indefinitely. We could not have staffed a team of speech-recognition researchers. Know which one you're building.
If the work needs to be in a room, check the room exists. Some categories need to sit next to their customers. Ours happens to be the opposite — our team's own neighbours are the customer — but that was luck as much as strategy.
What's left, once you subtract those, is a fairly specific profile: a small, bootstrapped, retention-sensitive company building for customers who look like the people already living there. That was us. It genuinely isn't everyone, and the cost of getting the match wrong is 18 months and a team that leaves.
If any of this sounds like a place you'd want to work, we're hiring. If you'd rather just see what the team built, start free with ₹500 of credit — or read more about us on the about page.