On June 30, 2026, AWS announced a $1 billion commitment to create a dedicated Forward Deployed Engineering (FDE) organization, led by Francessca Vasquez, VP of Frontier AI Engineering and Services. Engineers from the team will embed directly inside customer companies, working shoulder to shoulder with business, engineering, and security staff to build production-grade agentic AI systems, with the explicit goal of compressing deployments that once took months into days.
The timing is what makes the move notable: OpenAI and Anthropic each announced their own push into enterprise AI services barely two months earlier. With this step, AWS signals that the cloud giant has also decided that “delivering outcomes,” not “selling tools,” is the next currency of enterprise AI.
What the FDE Model Is
The term was popularized by Palantir: an engineer from the vendor works on-site at the client while the system is being set up, responds directly to issues as they arise, reuses much of the underlying technology across deployments, and customizes it to each company’s workflows. The benefit for clients is speed to results and, ideally, capabilities that stay behind in their own teams. The cost for vendors is that they must staff a full bench of embedded engineers; at its core, this is a labor-intensive business.
When AI agents entered the enterprise, this old model suddenly came back into fashion. The reason is practical: most companies can buy models and rent GPUs, but what actually blocks them is fitting agents into existing processes, data, and governance. TechCrunch’s observation is that more and more companies are willing to pay someone to “install it for me,” and dedicated service organizations are forming to meet exactly that demand.
AWS’s $1 Billion Version
Per Amazon’s own announcement, the $1 billion is an internal resource commitment, not a joint venture and not an investment in the usual sense. That is the sharpest divergence from the OpenAI and Anthropic playbooks.
The organization has several distinctive mechanics. Frontier teams arrive alongside purpose-built agents and run an “AI-Driven Development Lifecycle,” in which agents accelerate every phase while human engineers verify and guide the work. Deployments build a semantic layer inside the customer’s own AWS account, connecting to enterprise data and publishing a governed, versioned knowledge graph, so domain expertise lives in the customer’s systems rather than in people who will eventually rotate off. Engagements are structured around shared business outcomes instead of billable hours, and customer engineers are deliberately progressed “from observers to co-builders to autonomous operators,” so that when the engagement ends, the client can run things alone. Security features include hardware-based isolation and end-to-end encryption, with data staying inside the customer’s governance framework.
Customers Already in Production
This org does not start from zero: AWS’s Generative AI Innovation Center had already worked with BMW, Jabil, and Lyft. The new organization names the Allen Institute, Cox Automotive, the NBA, the NFL, Ricoh, and Southwest Airlines among its customers. NFL CIO Gary Brantley said the partnership took fan-facing products NFL Fantasy AI and NFL IQ into production “in just weeks.”
AWS is also pulling partners into the effort: partners will contribute model and industry expertise, while AWS invests in partner training and tools. The target audience is organizations past the experimentation stage, especially regulated industries, financial services, and the public sector.
How It Differs from OpenAI and Anthropic
In May, OpenAI’s “The Deployment Company” set out to raise $4 billion at a $10 billion valuation, with backers including TPG, Brookfield, Advent, and Bain Capital. Anthropic, hours later, announced a joint venture with Blackstone, Hellman & Friedman, and Goldman Sachs, valued at $1.5 billion per the Wall Street Journal, with $300 million commitments each from Anthropic, Blackstone, and Hellman & Friedman. The shared logic: borrow private equity’s capital and turn portfolio companies into customers, opening a new enterprise sales channel.
AWS picked a third path: no external fundraising, no revenue share with private equity, just a direct bet on its own balance sheet and its existing enterprise customer base. By comparison, Amazon is betting on scale and integration; the FDE org naturally locks into AWS accounts and governance tooling. The two labs are betting on channel innovation. Which route wins is anyone’s guess, but all three agree on one thing: the next wave of enterprise AI revenue comes from services, not from subscriptions alone.
Sources
- Amazon launches new $1 billion FDE org — TechCrunch
- AWS is investing $1 billion in forward-deployed AI engineers — Amazon
- Anthropic and OpenAI are both launching joint ventures — TechCrunch
AI-assisted summary compiled from the sources above, reviewed by a human before publishing.