Microsoft Launches $2.5B 'Frontier Company' - Embedding 6,000 Experts Inside Customer Enterprises

On July 2, 2026, Microsoft announced Microsoft Frontier Company, a new operating business dedicated to helping enterprises deploy AI successfully¹. Rather than selling AI tools and stepping away, Microsoft will send its own experts into customer organizations to build AI systems together, based on measurable business outcomes — a departure from the traditional software business model.

According to the announcement, Microsoft is investing $2.5 billion in Frontier Company and embedding 6,000 industry and engineering experts at customer companies¹. The move extends Microsoft’s enterprise AI strategy, building on earlier investment in its Azure AI Foundry development platform for enterprise AI. These experts will co-design and co-innovate AI systems with customers, and continue improving them after deployment. Two days before the announcement, Amazon Web Services (AWS) had disclosed a $1 billion commitment to a similar AI deployment business², marking a wave of large investments by major cloud providers in hands-on AI implementation support.

How Frontier Company Works

Frontier Company is an organization for what Microsoft calls “Frontier Transformation” — supporting AI-driven enterprise change end to end¹. Judson Althoff, CEO of Microsoft Commercial Business and author of the announcement, described it as an organization that “will be the largest, most capable, outcome-driven engineering organization in the industry”².

What stands out is how far it leans into outcomes. Experts embed inside customer companies and build and improve AI systems at scale based on measurable business outcomes¹. Althoff said “This goes beyond what has been labeled as Forward-Deployed Engineering”² — positioning the effort as an expansion, in both scale and outcome accountability, of the FDE (Forward Deployed Engineering) approach spreading across the AI industry, in which vendor engineers work inside customer organizations.

The announcement is also explicit about customer data. Microsoft stated that “a customer’s IQ is protected. Their data, their IP, their competitive advantage — none of it is used to train models”¹ — a message aimed at enterprises concerned about how their data is handled during AI adoption.

Leadership and Partners

Rodrigo Kede Lima, who brings 30 years of industry experience, will serve as President of Frontier Company. For the past six years at Microsoft, he has led enterprise-wide transformations for large companies¹.

For external partnerships, the company will work with major consulting firms including Accenture, Capgemini, EY, KPMG, and PwC¹. Initial customers include global enterprises such as the London Stock Exchange Group (LSEG), Unilever, Land O’Lakes, and Novo Nordisk¹.

The choice to partner with consulting firms rather than compete against them is notable. The structure appears designed to combine Microsoft’s engineering capacity with the industry knowledge of the major consultancies, dividing the work of enterprise AI adoption between them.

A Wave of Big Bets on AI Deployment

This announcement is not an isolated move. On June 30, two days earlier, AWS announced a $1 billion internal commitment to its own AI deployment business², and both OpenAI and Anthropic are reported to have launched joint ventures along similar lines². Beyond the race over model and tool performance, the ability to turn AI into real results on the ground inside enterprises is emerging as a new competitive axis.

In this competition, Microsoft has the advantage of its existing customer base. TechCrunch notes that Microsoft has already deployed engineers to much of the Fortune 500, giving the new effort a significant head start². Microsoft has also served as a technology partner in OpenAI’s Stargate infrastructure project, reflecting a consistent investment strategy spanning AI infrastructure through deployment support.

What This Means for Enterprises

For companies adopting AI, this shift changes the procurement model. Beyond the traditional pattern of buying licenses and implementing them with in-house teams or system integrators, enterprises now have the option of vendor experts working inside the organization, sharing outcome metrics, and building together. Many companies struggle to turn AI adoption into results, and a model in which the vendor takes on outcome accountability could be one answer to that problem.

At the same time, the deeper vendor experts go into a company’s operations, the greater the dependence on that vendor becomes. While the data protection policy is clearly stated¹, balancing the strategy of building internal AI capabilities against borrowing external execution capacity will be a decision for leadership teams.

Details of Frontier Company are available in Microsoft’s official blog announcement. The TechCrunch article covering the broader AI deployment trend is also a useful reference.