Building a global tech company from Kochi

Starting as a lab

In 2017, I founded ElkanIO Research Labs in Kochi with a focus that felt unusual at the time: an AI and analytics lab aimed at hard, high‑concurrency problems. I was shaped by postgraduate work in AI and years of leading data teams. The common lesson was simple: if the underlying architecture is fragmented, no amount of individual brilliance saves the system from drift. The unit of progress is the model the business runs on.

Kochi was a deliberate choice. It put me close to the talent I wanted to grow with, and it kept me honest about what matters. Building from Kerala is not a constraint; it’s a forcing function. You build clear systems or you spend your days fighting handoffs. I chose the first path.

From research to platform

The lab work pulled me into travel technology. The challenge wasn’t ambition; it was a stack that asked people to live in tabs, re‑key data, and react to supplier changes as incidents. I saw the same “patchwork mentality” across teams: a dozen tools stitched together with brittle integrations. Every new feature became a one‑off project. Margins eroded in the background as work moved between silos.

Our response was to turn research into infrastructure. We built a core we call Synapse: a unified schema and API that represent the truth once across verticals—flights, hotels, activities, insurance, visas, cabs, ferries, cruises. Around that core we built products for real use cases: Nucleus for B2B distribution, Nova for B2C, and more to connect operations, customer, and finance.

The goal wasn’t a single tool that does everything. It was one architecture that many products can trust. That choice—one schema, many surfaces—changed the work. We stopped solving the same problem in five places. We started compounding improvements.

Architecture beats proximity

Founders ask me what it’s like to build from Kochi. My answer is consistent: location sets the constraints; architecture sets the ceiling. If your model is clear and your contracts are explicit, you can ship from anywhere. If your system is fragmented, moving to a new postcode just adds another hop.

I’ve seen the benefits up close:

• Talent compounding: Kerala has engineers who care about systems. When the architecture is coherent, their work stacks instead of scattering.
• Design pressure: Distance from hype forces better fundamentals—observability, consistency, and clear responsibilities.
• Global benchmark: When you build for travel, you build for the world by default. The platform’s bar is the same whether a user sits in Kochi or New York.

Agentic automation as a habit

AI in travel is most valuable when it closes loops. We design around a simple rhythm that runs through the platform:

watch → propose → prove → promote → route

• watch: instrument flows and spot drift
• propose: generate a candidate mapping or behavior change
• prove: validate safely before rollout
• promote: ship with guardrails and a way back
• route: choose suppliers with intent, not guesswork

This agentic loop is not a feature; it’s a habit. It only works because the core is stable. Synapse gives automation something coherent to learn from. That’s how we scale without throwing more people at exceptions.

Lessons for founders building from India

If you’re starting or scaling from India—Kerala or anywhere—the hardest and most valuable choice is to treat architecture as your primary product. A few things I’ve learned:

1. Own the model. Represent truth once. The number of tools you plug in matters far less than the quality of the schema they plug into.
2. Put analytics inside the system. Yield engineering isn’t a dashboard; it’s how the platform discovers margin across routing and operations.
3. Make change safe. Explicit contracts between products (B2B, B2C, corporate, groups) beat ad‑hoc glue every time.
4. Write the boring parts. Observability and reconciliation are not chores; they are capabilities.

This isn’t romantic. It’s practical. Architecture is how a small, persistent team can build something global from Kochi.

Kerala talent, global outcomes

I’ve had the privilege of building with engineers who value depth and clarity. The multiplier is not a trendy headquarters; it’s a system that lets good decisions land and stay. When you give a strong team a coherent core, they move faster without breaking the truth.

If you’re curious about the platform choices behind this, I’ve written about Synapse and why we chose a single schema for travel: Building Synapse: one schema for travel. And if you want the context on the problem we set out to solve, start with The patchwork mentality in travel tech.

For a view into my background and focus areas, the About page has more.

Closing

We started as an AI lab in 2017. We’re now building a travel technology ecosystem from Kochi that treats architecture—not headcount—as the advantage. If you’re a founder, operator, or builder working on similar problems, I’d love to connect.

Get in touch →