Google Cloud and Accenture are collaborating to create a dedicated unit to help enterprises adopt Google’s AI tools and services by sending engineers to work directly with companies and design custom applications suited to their operations. The unit is called the Accenture Gemini Enterprise Business Group and will operate under Accenture, according to a statement from Google.
Google plans to train up to 1,000 field deployment engineers (FDEs) affiliated with Accenture to work with enterprises and build custom applications on the Gemini Enterprise platform. This model is emerging at a time when AI and cloud-computing companies are beginning to view direct assistance with solution implementation as a standalone commercial phase, rather than merely an accompanying service for selling AI models.
A Race to Overcome the Implementation Bottleneck
Enterprises are struggling to integrate AI tools into workflows in ways that deliver tangible savings or revenue. The field deployment engineer model is betting on providing teams that combine an understanding of business needs with the ability to build solutions based on agentic AI, instead of leaving the enterprise to handle the integration process on its own.
The competition is not limited to Google. OpenAI, Anthropic, Microsoft, and Amazon have launched similar units or programs, while companies specializing in building custom workflows, such as Anthropic-linked Ode and OpenAI’s The Deployment Co., are competing with major consulting firms in this field.
An Attempt to Strengthen Google’s Enterprise Presence
The partnership comes as part of Google’s expansion of the field deployment engineer model. Earlier this year, Google Cloud announced a commitment of $750 million to a partner ecosystem that includes embedding its engineers within consulting firms such as Capgemini, Cognizant, and Deloitte. It also entered into a multiyear partnership with CVC Capital Partners to deploy these engineers within companies in its portfolio.
According to August data from Ramp, Google accounted for approximately 6% of enterprise AI spending among Ramp’s U.S. customers, compared with 43.5% for Anthropic and 39.7% for OpenAI. Google explained that these data may not include many large enterprises with which it has strategic deals extending beyond the use of model APIs, such as Oracle, Meta, Anthropic, and ServiceNow.
certi.news Analysis: What Is Actually Changing?
The most important development here is not the launch of a new product, but the transfer of a larger share of responsibility for AI success to implementation teams working inside enterprises. This could help Google turn demand for Gemini from experiments and APIs into customized operational projects, but it does not by itself prove that these projects will generate sufficient returns.
This issue is particularly important because Google Cloud generated revenue of $24.8 billion in the second quarter, while parent company Alphabet accumulated contractual obligations and purchase commitments totaling, according to the article, $811 billion as of June 30. As massive investments continue to flow into graphics processing units, data centers, and energy, the open question remains whether deploying engineers inside companies will actually remove adoption bottlenecks or add another consulting layer without guaranteeing a clear return on customers’ spending.