Infor cuts automation prototyping to weeks with forward-deployed engineers

Infor's argument, as laid out by CEO Kevin Samuelson and President and CTO Soma Somasundaram at Infor Velocity Week, is that the company spent years building software around the specific processes its customers use in their industries. That depth, combined with an open architecture, is what Samuelson says let Infor put AI on top of existing workflows to deliver outcomes rather than demos.

The timing reflects a gap the company's own global research points to: two in three businesses said off-the-shelf AI doesn't adequately address their industry's needs. Somasundaram's example of an automotive or parts supplier, where material shortage is not an option, illustrates why generic tools struggle — the value sits in process detail, not model capability alone. Infor Industry AI is positioned as an architecture embedded across its applications on the Infor OS foundation rather than a standalone product.

The trust question is handled in stages, starting with a human reviewing agent recommendations and adding observability, auditability and access controls as agents take on more work. Whether the one-to-three-week prototyping timeline holds as use cases grow more complex is likely the test for buyers; the forward-deployed engineers are the mechanism, but they are also a cost Infor carries, and the interview does not say how that model scales.