Microsoft Launches Microsoft Frontier Company, Investing $2.5 Billion and Deploying 6,000 Experts On-Site with Customers for AI Transformation
- Microsoft has formed a new business unit, Microsoft Frontier Company, investing $2.5 billion and deploying 6,000 industry and engineering experts inside customer organizations to drive "Frontier Transformation."
- Microsoft positions this playbook as going beyond the existing "Forward Deployed Engineering" (FDE) approach, with the core method being a continuous improvement loop that never stops running between two platforms: Intelligence and Trust.
- Microsoft's commitment: customer data, IP, and competitive advantage won't be used to train models in ways that erode their differentiation; the platform supports freely switching between OpenAI, Anthropic, Microsoft's own models, or open-source models.
- Already deployed at London Stock Exchange Group (LSEG), embedding AI into LSEG Workspace to help financial professionals query content; also names customers like Land O'Lakes, Unilever, and Novo Nordisk, though without elaboration.
- The unit is led by President Rodrigo Kede Lima (30 years of industry experience, 6 years in Microsoft sales), and will scale up jointly with consulting partners including Accenture, Capgemini, EY, KPMG, and PwC.
Microsoft is standing up a dedicated team specifically to help enterprises turn AI into real, measurable value
Judson Althoff, CEO of Microsoft's commercial business, announced on the official Microsoft blog on July 2, 2026, that Microsoft will launch a new business unit, Microsoft Frontier Company, joining forces with partners to drive "Frontier Transformation" for enterprise customers worldwide.
In plain terms, Microsoft is pulling together a dedicated team: investing $2.5 billion and placing 6,000 engineers and industry experts directly inside customer companies, working alongside them to design, deploy, and continuously refine AI systems—with one goal: business results that can actually be measured.
Why customers need this now: money spent has to show results, and expertise can't be siphoned off by AI
Microsoft's read on the situation is that enterprises are well past the "let's just try it" phase. What customers want now is measurable business results from the AI budget they've spent—proof the investment was worth it.
At the same time, enterprises are carrying a lingering concern. The proprietary data, workflows, and industry expertise they've built up over years is what sets them apart from competitors. They worry that once this gets absorbed into general-purpose models by AI, it becomes an ordinary capability anyone can call up—flattening their own moat.
Microsoft distills these two concerns into two words: Intelligence (amplifying yours) and Trust (making it trustworthy). Judson Althoff has written previously that these are the two most important pieces of any AI solution, and they form the foundation for everything the organization is building on top.
Two pieces of the puzzle, viewed separately first: one accumulates your proprietary edge, the other keeps a close watch on your AI systems
Achieving both "amplified intelligence" and "trustworthiness" at the same time, Microsoft says, requires two independent platforms underpinning it. They handle different jobs—let's look at each on its own first.
Capturing your unique know-how
- Captures proprietary enterprise data, domain expertise, workflows, and decision processes
- These capabilities compound over time the more they're used inside the enterprise
- Freely choose models when building solutions, without being locked to any single one
Seeing clearly into your AI systems and keeping them under control
- Covers every layer of the tech stack—observability, governance, management, and security for AI systems
- Uses FinOpsFinOps: a methodology for accounting for and managing cloud computing and AI spend, mapping every dollar invested to the actual business value it delivers—making it easy to judge whether the money was well spent. to account for the return on this investment
- Lets the enterprise clearly see how well its AI is actually performing and whether it's worth it
What the 6,000 experts actually do: keep these two pieces of the puzzle turning, continuously
A typical AI consulting engagement ships the system and moves on. Microsoft Frontier Company keeps engineering experts embedded on-site, building a continuous improvement loop between the Intelligence platform and the Trust platform—repeatedly fine-tuning agentic business processes so the customer's intelligence compounds over time, eventually landing real business results. This is what sets it apart from typical AI consulting, and from typical FDE.
The core job of the 6,000 experts is to keep information flowing continuously between those two platforms above. Data and expertise feed in, the system produces results, the results flow back to the experts for optimization—round after round, on and on.
The "agentic business processes" mentioned here refers to processes that can carry out multiple steps on their own, deciding for themselves which tools to call to get the job done, without needing a human to click through every step. The experts' job is to make these processes get more precise with every round of feedback.
The on-site engineers driving this loop use an approach called "Forward Deployed Engineering" (FDE): the tech company places engineers directly inside the customer's organization, where they write code, deploy systems, and solve real problems alongside the customer's own team. It's a bit like an appliance maker that doesn't just sell you the appliance—they also send an engineer to live at your house and keep tuning it until it works perfectly.
Written into the terms: your data won't be used to feed a model that your competitors can use too
Microsoft calls one principle "non-negotiable": the customer's intelligence is protected. Your data, your IP, your competitive advantage will never be used to train a model in a way that "erodes your industry differentiation." The way this is protected is through an open, model-agnostic platform that lets enterprises pick the right model for each scenario, without being locked to any single vendor.
For an enterprise, this commitment is what decides whether they're willing to hand over their crown jewels. Microsoft lays out the difference between the two approaches quite clearly.
- Your proprietary expertise gets absorbed and turned into a generic capability anyone can call on
- Industry differentiation gets "commoditized," flattening your moat
- Your solution is locked to a single model, a single vendor, with no way to switch
- Data, IP, and competitive advantage never enter training—they stay in your own hands
- Freely choose models by scenario: OpenAI / Anthropic / Microsoft's own / open-source / industry-specific models
- No vendor lock-in—procurement decisions stay entirely in your own hands
There's no social consensus that would allow an AI's future to consume the very intelligence of the companies that deploy it.Satya Nadella, Microsoft CEO (quoted by Judson Althoff)
Building on this principle, Microsoft's solution is an open, heterogeneous multi-model platform: enterprises shouldn't be locked to a single model, just as they shouldn't be locked to a single technology vendor. Within the same system, each scenario can run whichever model fits it best, with control never handed entirely to any one provider.
Who's already using it: at the London Stock Exchange, analysts ask AI directly for answers
Of the deployment examples Microsoft gives, only London Stock Exchange Group (LSEG) is described with any specific mechanism—the rest are just name-dropped.
Microsoft's engineers and industry experts worked with LSEG to embed AI into its LSEG Workspace, letting financial professionals ask complex questions directly against structured and unstructured financial content and get fast answers. This is continuously refined under the hood using customer feedback and real-time user testing, with each iteration getting faster and the model's quality and coverage improving bit by bit.
Beyond LSEG, Microsoft also names customers like Land O'Lakes, Unilever, and Novo Nordisk, along with global consulting partners including Accenture, Capgemini, EY, KPMG, and PwC, saying it will rely on these partners to roll this model out across global markets and industry verticals. The official blog post doesn't go into specifics on any of these examples.
Who's steering the ship—putting the numbers side by side
Steering this new unit is Rodrigo Kede Lima, taking on the role of President. He brings 30 years of industry experience, having spent the past 6 years at Microsoft leading sales, running enterprise-scale transformation across the Americas and Asia, and consistently helping customers and partners turn technology shifts into business outcomes.
At the end of the day, it comes down to two words: Intelligence and Trust—helping customers achieve meaningful outcomes and get a return on their investment.Judson Althoff, CEO of Microsoft's Commercial Business