Neysa at
Data & Gen AI Summit 2026, Delhi

Thank you for meeting us in Delhi

It was wonderful to meet so many of you at the Data & Gen AI Summit in Gurgaon. The event only lasted a day, but thank you for being so open about where your AI work is today and where you hope to take it next. Those honest conversations are exactly why we love being part of this community.

One thing became clear as we spoke: while everyone is moving at incredible speed, there’s a shared focus on making sure the foundations are strong enough to support that growth. That sense of forward momentum really set the tone for the entire day.

Happenings at the booth

The booth ran all day, and most of these conversations happened standing up, over someone’s architecture sketched on a notepad. We walked people through our AI platform Neysa Velocis.

But the questions that took us longest to answer were rarely about GPUs. They were about who can see what, where the data sits, and what the invoice looks like at the end of the month.

Neysa on Stage

The conversation continued on the main stage, where our Chief Revenue Officer, Xavier Kurian, shared his perspective on the future of AI infrastructure.

Xavier spoke about the shift from training models to serving them at scale. While training is a one-time event, serving users happens around the clock and requires a different kind of budget and infrastructure. He explained how purpose-built layers are now appearing alongside general-purpose clouds to help teams manage these production demands without starting from scratch.

The question that kept coming up

• Getting something working is not the problem. Nearly everyone we met had a pilot that worked. Far fewer had it live. And when we asked what was holding it up, the model was almost never the reason.
• The bill turns up before the value does. Pricing that felt perfectly manageable during a pilot stops feeling that way the moment a workload runs around the clock. Several people told us the finance conversation started later than it should have.
• Governance shows up last and stops everything. Residency, audit and access requirements tend to land right at the production gate, which is the most expensive possible place to answer them.