Dive Brief:
- Microsoft is spending $2.5 billion to launch a new operating business called the Microsoft Frontier Company, which will embed 6,000 engineering experts within customer operations to assist in deploying AI systems at scale, according to a Thursday blog post written by Judson Althoff, CEO of Microsoft Commercial Business.
- The unit will focus on end-to-end frontier transformation, with customers including Land O’Lakes and Unilever already taking advantage of the service, Althoff said in the blog. Microsoft’s announcement comes two days after AWS shared its $1 billion investment in a forward deployed engineering organization.
- “To achieve scale, we will work closely with our partner ecosystem to extend this unique value to our customers across all markets and segments globally,” Althoff said, adding that Microsoft has FDE partnerships with Accenture, Capgemini, EY, KPMG, PwC and others in its partner ecosystem to help expand the effort.
Dive Insight:
Microsoft and AWS’ near simultaneous launch of organizations dedicated to embedding thousands of engineers within customer operations reflects a shift in approach to AI delivery among tech providers.
By the end of the year, 85% of tech providers will have established FDE programs as core AI delivery models, Alex Coqueiro, senior director analyst at Gartner, told CIO Dive in an email. Competition for the scarce skillset is already intensifying, he added. Google Cloud is also on an engineer hiring spree for its AI-focused go-to-market team.
“Enterprises want FDEs because they compress deployment timelines and close the ‘last mile’ gap between pilot and production, something standard consulting and off-the-shelf tools cannot do in complex environments,” Coqueiro said.
Yet CIOs will need to pause and think strategically before tapping into such offerings, Coqueiro said.
In the next two years, 70% of enterprises will be forced to abandon agentic AI projects springing from FDE-led engagements due to both high vendor costs and lack of internal skills to keep developing the technology independently, according to an April report from Gartner.
FDE engagements operate on compressed timelines, which AWS calls “45-day sprints,” according to Francessca Vasquez, VP of frontier AI engineering and services at AWS.
Gartner estimations suggest FDE consulting fees before platform and integration costs could range between $200,000 and $400,000 quarterly per use case.
Without a strategic plan, executing a project with FDEs could leave an organization with a bespoke product that requires vendor maintenance, creating dependency, according to the Gartner report.
“Without a clear exit plan and internal ownership, FDEs quietly become permanent staff augmentation, driving vendor lock-in and eroding internal AI capability,” Coqueiro said.
Enterprise technology leaders should take multiple steps before and during an FDE engagement to ensure success, the Gartner report said. That includes selecting the right problem — such as an operationally complex bottleneck — for an FDE engineer to solve. CIOs should then estimate the full integration burden before committing to the engagement.
Next, engineers must be paired with enterprise domain experts who are positioned as co-designers of the new business workflow, with weekly working increments to keep projects grounded. Contracts and holding vendors accountable through product delivery, development and handoff will also be critical to long-term success, the Gartner report said.
In their announcements this week, both Microsoft and AWS emphasized enabling customers to operate independently after engineers depart.
Althoff said the Microsoft Frontier Company aims to help enterprises establish an intelligence platform that protects their decision-making and proprietary data while tapping into their choice of AI models to build products and workflows, as well as a trust platform for governing and managing AI across the tech stack.
“Enterprise AI engineering expertise with deep industry knowledge is required to build a system that acts as a continuous loop of improvement between the two platforms to fine tune agentic business processes, ensuring that a customer’s intelligence compounds over time and delivers real business outcomes,” Althoff said.
