Amazon has launched a new $1 billion organization called FDE, short for Field Deployment and Enablement. The new unit will embed Amazon engineers directly inside client companies to build and deploy purpose-built AI agents.
The move follows a similar strategy already adopted by OpenAI and Anthropic, two of the leading AI companies that Amazon has heavily invested in.
What is the new Amazon FDE org?
The FDE org is designed to help Amazon's customers move faster in adopting AI agents. Instead of just providing cloud infrastructure or AI models, Amazon will now send its own engineers to work inside client companies. These engineers will build custom AI agents tailored to each client's specific business needs.
The focus is on speed. Amazon wants these agents deployed quickly, and the company aims to make customers self-sufficient so they do not need ongoing support from Amazon after the initial deployment.
Why Amazon is following OpenAI and Anthropic
OpenAI and Anthropic have already been embedding their engineers inside large enterprise clients to help them deploy AI agents. Amazon's new FDE org is a direct response to this trend. By putting engineers on the ground, Amazon hopes to win more enterprise customers for its AI services, particularly Amazon Web Services (AWS).
This approach is different from Amazon's traditional model of selling cloud services and letting customers figure out the implementation themselves. Now, Amazon is taking a hands-on role to ensure customers can actually use AI agents effectively.
Our Take: A smart but risky bet
In our view, Amazon's FDE org is a smart strategic move. The company has invested billions in AI startups like Anthropic and OpenAI, but it has struggled to build its own frontier AI models. By focusing on deployment and enablement, Amazon is betting that its strength lies in helping customers use AI, not in building the best AI model itself.
However, this approach is expensive. Embedding engineers inside client companies costs a lot of money and talent. The $1 billion investment shows Amazon is serious, but it also raises questions about whether this model can scale. If customers become truly self-sufficient, Amazon may need fewer engineers over time. If they do not, the company could be stuck with a high-cost service model.
For now, Amazon is following the playbook of OpenAI and Anthropic, but with its own twist — using its massive cloud infrastructure and engineering talent to win the AI deployment race.
