AWS has launched a new Forward Deployed Engineering (FDE) organisation backed by a US$1bn investment, embedding expert engineers directly within customer teams to help enterprises move beyond experimental AI projects and into production-ready, agentic AI systems.
How the FDE Model Works
The FDE model operates on three principles: it prioritises agentic AI, it compresses deployment timelines from months to days, and it aims to leave customers capable of independent operation once the engagement concludes.
Engineering teams use what AWS calls an AI-Driven Development Lifecycle, combining AI-powered agents with human oversight. Rather than simply bolting AI tools onto existing workflows, the approach integrates agents throughout each stage of development while engineers validate and oversee the process.
"Customers have moved past exploring what AI can do; they want to make it core to how they operate."— Francessca Vasquez, Vice President of Frontier AI Engineering and Services, AWS
Security and Governance Considerations
Security has been embedded into the FDE framework from its inception. According to AWS, customer data remains within existing governance structures, while hardware-based isolation and end-to-end encryption provide additional layers of protection for enterprise deployments.
This approach could prove particularly relevant for organisations operating in highly regulated sectors such as financial services and government.
Early Adopters Report Results
Several major organisations have already engaged with AWS Forward Deployed Engineering, including the Allen Institute, Cox Automotive, the NBA, Ricoh, Southwest Airlines and the National Football League.
Gary Brantley, Chief Information Officer of the NFL, said the league partnered with AWS FDE to get engineers building alongside its team, launching new products into production in just weeks — including fan-facing tools such as NFL Fantasy AI and NFL IQ.