Full story
AWS bets $1B on FDE
Amazon Web Services announced it is investing $1 billion to create a Forward Deployed Engineering organization that will embed thousands of engineers with enterprise customers to build and deploy AI agent systems.
“AWS is investing $1 billion in a new Forward Deployed Engineering org”
The FDE model will send small pods of roughly five or six engineers into client companies on a “45-45-45” cycle, with 45 minutes to ideate, 45 hours to validate, and 45 days to ship production solutions.
AWS said the approach is “agentic-first,” with pods deploying purpose-built AI agents alongside human engineers, and it is designed so customers are fully self-sufficient when a deployment ends.
The announcement also places AWS alongside OpenAI and Anthropic, which launched their own enterprise deployment units in May 2026, with OpenAI’s Deployment Company raising over $4 billion and Anthropic’s services venture valued at $1.5 billion.
Early customers named by AWS include the NFL, NBA, Southwest Airlines, Cox Automotive, Ricoh, and the Allen Institute, as the company positions the new unit as a hands-on push into real-world AI implementation.
Vasquez: agentic-first, speed
AWS VP of frontier AI engineering and services Francessca Vasquez told CIO Dive that the FDE model is designed to be “agentic-first,” with pods of five to six human engineers working inside a customer’s organization initially while AI agents continue running long-term.
Vasquez also described the effort as “rewiring our last mile of deployment,” and she said AWS FDE targets end-to-end business workflows rather than billable hours.
About AmazonIn an AWS blog post, Vasquez said the organization is designed so “Customers leave AWS FDE deployments with both new solutions and new engineering capabilities,” and she tied that to leaving behind knowledge graphs, runbooks, architectural documentation, and trained internal champions.
CIO Dive reported that a semantic layer will be deployed into customers’ AWS accounts to connect to enterprise data sources and publish a structured knowledge graph that AI agents can reason from.
SiliconANGLE added that Vasquez said the “new currency of value is speed,” and it quoted her 45/45/45 metric as “Ideate on in 45 minutes, validate that idea in 45 hours, and then ship something within your workflow of value in 45 days.”
Who’s using it, what’s next
AWS said customers including the Allen Institute, Cox Automotive, the NBA, the NFL, Ricoh, and Southwest Airlines are already working with AWS FDE teams, as the company frames the work as moving from AI experimentation to production.
“Customers leave AWS FDE deployments with both new solutions and new engineering capabilities”
About AmazonGary Brantley, the NFL’s chief information officer, said the league worked with AWS FDE to “innovate at the pace and scale needed to meet the high expectations of our fans,” and he linked the partnership to launching fan-facing products including NFL Fantasy AI and NFL IQ.
The AWS blog post described how FDE deployments are structured around shared goals and business results, and it said the semantic layer is deployed into the customer’s own AWS account to publish a governed, versioned knowledge graph.
CIO Dive reported that AWS FDE is part of a broader wave of FDE programs, noting that Google Cloud listed job postings for 59 engineering roles for an AI-focused unit and that Accenture shared plans with Microsoft to launch a forward deployed engineering practice.
AWS’s $1 billion commitment also comes as the term “forward deployed engineer” is described as having been coined by Palantir more than a decade ago and as job postings for the role surged more than 800% between January and September 2025, underscoring the competitive push to embed AI teams inside enterprises.