Our Story

Vajra started with a problem we couldn't stop thinking about.

For years, we worked close to the operational side of financial markets. We saw how much of the industry depended on reconciliation and how much of that work still happened the same way it had for years.

A break would be identified. Then someone had to investigate it. They would pull reports from different systems, search through emails, open PDFs, compare transactions, trace balances back through the workflow and piece together what actually happened.

The technology could tell you that something was wrong. But far too often, it couldn't tell you why or what to do about it. And the more complex the instrument, the messier the data, or the more systems involved, the more manual the investigation became.

We kept asking the same question. If technology can identify the exception, why can't it help resolve it?

That question stayed with us. The industry had spent years building increasingly sophisticated systems for matching and exception detection. But the last mile, understanding the break, gathering the evidence, determining the cause and figuring out the next action, was still largely dependent on people.

We believed AI could change that. Not by simply adding a chatbot to an existing workflow. But by building something that could actually reason across the data, documents and context that operations teams use every day.

That's where Vajra began.

We started building Vajra around a simple idea: Reconciliation is not a matching problem. It's a resolution problem.

Vajra brings together the information surrounding an exception and uses AI to investigate what happened, determine why records don't agree and help identify the appropriate resolution.

It is designed for the reality of financial operations, where data isn't always clean, information isn't always structured and the answer is often buried across multiple systems and documents.

We are building from experience, not theory. We've lived with these workflows. We've seen the spreadsheets, the reports, the emails, the PDFs and the endless back and forth.

We know that a break isn't just a red flag on a screen. Someone has to resolve it.

That is the problem we wanted to solve.

And that is why we built Vajra. Not to make it easier to find exceptions. To make it possible to resolve them.