AWS commits $1B to embedded AI unit

Amazon Web Services has committed $1 billion to establish a new Forward Deployed Engineering (FDE) organization focused on artificial intelligence. The new unit is designed to embed AWS engineers directly with customer teams to build and deploy AI systems in production. AWS announced the initiative at its Summit event in Washington, D.C., positioning the program as a way to accelerate customers’ AI projects.

The FDE organization will be seeded with thousands of engineers, indicating a large-scale effort to place specialized AI talent inside customer environments. These engineers will focus on practical deployment work rather than purely advisory roles, working closely with business, engineering and security staff at client organizations.

AWS describes the FDE engagements as running in roughly 45-day cycles. For each engagement, a pod of about five or six engineers will be embedded within a customer at a time. This structure is intended to deliver tangible results within weeks rather than over long, open-ended projects.

During these cycles, FDE teams partner with customers on defined business outcomes instead of traditional project-based or hourly models. The goal is to leave behind self-sufficient teams and new capabilities when the engagement ends, rather than creating long-term external dependencies.

A central element of the FDE approach is the use of agentic AI and purpose-built AI agents. These tools are designed to complete tasks independently on behalf of users, supporting the embedded engineers as they work through customer workflows. AWS says this combination helps compress development timelines from months down to days.

FDE teams will work alongside these AI agents in customers’ environments to integrate agentic AI patterns directly into business processes. By focusing on rapid deployment and iterative delivery, the teams aim to achieve faster value realization than traditional, project-based engagements.

AWS has already identified several organizations working with FDE teams. These include the Allen Institute, Cox Automotive, the National Basketball Association (NBA), the National Football League (NFL), Ricoh and Southwest Airlines (LUV). The list spans scientific research, automotive services, professional sports, imaging and office solutions, and air travel.

The diversity of these initial customers highlights that the FDE program is not limited to a single sector or use case. Instead, it targets a range of industries looking to integrate AI more deeply into their operations, from analytics and decision support to operational workflows.

To build out the FDE unit, AWS plans to both hire new staff and reassign existing employees into the organization. This mixed staffing approach supports the goal of seeding the group with thousands of forward-deployed engineers. It also indicates AWS is reallocating internal resources to prioritize embedded AI work.

By combining a significant financial commitment with a large engineering footprint, AWS is positioning the FDE organization as a key part of its AI services strategy. The model focuses on close, time-bound collaborations that aim to rapidly deliver AI capabilities while enabling customers to maintain and extend those solutions on their own.