Share with your CIO
Kinetic IT is betting that the advisory-to-deployment gap is where AI projects go to die. The Australian managed services firm has launched a dedicated AI Engineering and Platforms division, appointing Marcio Sete to lead it, and is recruiting four Forward Deployed Engineers to embed directly inside client operations across aviation, utilities, government, and resources. The division sits under the Chief Transformation Officer rather than in a standalone technical practice, a structural choice that signals Kinetic IT sees AI as an integration problem first and a technology problem second.
What this means for your business
The proof-of-concept graveyard is a real place. Any CIO who has watched a successful AI pilot stall at the handoff to production knows the pattern: the consulting team that built the demo doesn’t own the operational environment, and the internal team that owns the environment doesn’t have the AI depth to carry it forward. Kinetic IT is making a structural argument that the only way to close that gap is to stop treating those as two separate engagements.
The Forward Deployed Engineer model, where senior technical staff live inside the client’s operational environment rather than visit it, is worth taking seriously as a procurement signal. Palantir built its enterprise footprint on exactly this approach, and the durability of those customer relationships came precisely from engineers who understood both the software and the operational constraints of a refinery or a logistics hub. When your AI vendor’s engineer knows your SCADA system layout by heart, the switching cost rises substantially. That’s a feature for the vendor and a tradeoff worth naming before you sign.
The question worth holding: as more system integrators launch embedded engineering models, the competitive pressure on internal AI teams intensifies in a specific way. It’s not that the vendor replaces your team. It’s that your team gets benchmarked daily against engineers who do this across five clients simultaneously. The signal worth watching is whether Kinetic IT’s four initial engineers scale to forty, which would indicate genuine market pull, or plateau, which would suggest the embedded model is harder to staff and margin than it looks on paper.
Concept deep-dive: Forward Deployed Engineering
Forward Deployed Engineering places product or AI engineers directly inside a client’s operational environment for an extended period, not as consultants who deliver recommendations, but as builders who write production code alongside the client’s own team. It exists because complex operational environments, think air traffic management systems or utility SCADA networks, have constraints that no amount of documentation captures. The analogy is the difference between designing a hospital layout from blueprints versus spending six months working a shift. For CIOs evaluating AI deployment partners, it’s the difference between a vendor who finishes the project and one who finishes the integration.
Based on reporting from Kinetic IT launches AI engineering division with Sete, originally published 2026-07-07 01:43:00.
