Microsoft has launched a new business unit that will focus on helping companies deploy artificial intelligence at scale, committing $2.5 billion and assigning about 6,000 staff to work directly inside customer organisations.
The AI deployment unit, called Microsoft Frontier Company, will bring together engineers, consultants, sales staff and industry specialists.
They will sit with clients, design systems with them, and support ongoing AI deployment inside business operations.
Judson Althoff, who leads Microsoft’s commercial business, said the company is building something larger than typical deployment teams.
“This goes beyond what has been labelled as Forward-Deployed Engineering (FDE),” he said, “and will be the largest, most capable, outcome-driven engineering organisation in the industry.”
The model is not entirely new in the tech sector as Amazon Web Services recently committed about $1 billion to a similar structure focused on AI deployments.
Other AI firms have also moved in the same direction, with OpenAI and Anthropic both setting up forward-deployed engineering teams in recent months, usually working with consultants and private capital partners.
Microsoft’s approach leans heavily on embedding staff inside large enterprises already using its cloud systems. The company says this gives it a head start in industries where it already has strong relationships.
Early deployments include work with financial data provider London Stock Exchange Group, consumer goods firm Unilever, and food company Land O’Lakes. Microsoft also mentioned healthcare company Novo Nordisk among early adopters.
At the London Stock Exchange Group, Microsoft engineers helped build AI features into its Workspace platform. The system allows users to ask complex questions and receive answers drawn from both structured and unstructured financial data.
Microsoft says the system continues to improve as users interact with it.
Althoff also noted that customers are now asking more questions about how AI should fit into their operations. He said firms are no longer just asking what the tools can do, but how to structure their entire workflows around them.
He also pointed to challenges in the market. Companies are still deciding whether to commit to a single model provider or mix several systems depending on use cases.
Microsoft argues for flexibility, saying customers should be able to use different AI models depending on need, without locking themselves into one ecosystem.
The company described its platform as “model-diverse” and said customers should not be tied to one vendor. Instead, organisations should be able to choose between models from OpenAI, Anthropic, Microsoft’s own systems, open-source options, or industry-specific tools.
Placing data control at the forefront, Microsoft says customer information will not be used in ways that weaken competitive advantage.
“There is no societal permission for an AI future that eats the intelligence of the companies it’s deployed inside,” Satya Nadella said.
Microsoft says this principle affects how the new unit will operate, customer data and intellectual property will remain protected and separate from model training in ways that could expose proprietary information.
The company has long worked with enterprise clients through consulting and support services, but it says this new structure goes further. It combines engineering, industry knowledge, and continuous deployment in one operating model.
Rodrigo Kede Lima will lead the new unit, having spent three decades in the technology sector and previously led Microsoft’s business across Asia as well as parts of the Americas.
He has also worked closely with enterprise customers on large-scale digital transformation projects.
Microsoft has been investing heavily in artificial intelligence infrastructure and has built large data centre capacity, rolling out products such as Microsoft 365 Copilot and GitHub Copilot, though adoption has varied across markets.
The tech giant has not hidden the uneven pace of uptake. Some AI tools have spread fast, while others have found it difficult to gain broad use in enterprise settings.
Companies are spending heavily on AI infrastructure, but many are still working out how to turn that investment into steady returns.
Microsoft’s commercial services generated about $2.1 billion in the March quarter, growing slightly from the previous year. The company says its strongest results come when it works closely with clients to build what it calls an “intelligence platform” around their existing systems.
That means helping firms connect data, manage models, and track performance across business units.
Microsoft says the goal of establishing the unit is not just AI deployment but ongoing adjustment, where systems are refined based on how they perform inside real operations.
