On Thursday, Google Cloud CEO Thomas Kurian issued a call for “forward-deployed engineers” to apply for jobs in the company’s go-to-market AI team. Their task: help non-tech organizations scale up their AI deployments.

That term — forward-deployed engineers, FDE for short — has been coming up a lot lately in conversations with CTOs, software engineers, and experts tracking the technology and job markets.

Google currently has 1,513 openings for that specific role and OpenAI, which just this week launched an organization called the Deployment Company, has 31. Microsoft is on board, too; in March, it partnered with Accenture to launch a forward-deployment partnership.

OpenAI’s new Deployment Company is, not surprisingly, designed to “help organizations build and deploy AI systems they can rely on every day across their most important work,” the company said in a blog post.

Forward-deployed engineering has seen the fastest growth in jobs created by AI, with the number of positions increasing 42-fold between 2023 and 2025, LinkedIn reported in a study earlier this year. (AI engineer jobs, by comparison, have grown 13-fold in that same time frame.)

Vendors and service providers created the FDE position to help clients install AI, said Jack Gold, principal analyst at J.Gold Associates.

Many non-tech firms have taken shots at deploying AI projects internally, without success or quick ROI. Some of the reasons those efforts haven’t worked out include poor vision, lack of talent, skimpy budgets, and underestimating the complexity of deploying AI.

That’s led to the arrival of FDEs — essentially hired guns for AI deployments. They focus on successful outcomes for customers instead of writing code. 

“They have skills that the organization may not have, and usually have done similar work with others before, so they bring expertise that companies need,” Gold said. 

FDEs analyze strategies, battle plan, discover applications, build agentic frameworks, and roll out AI systems with help from customers’ own domain experts and engineers. They also work with AI models, solve context and reasoning problems, evaluate models, and put security and governance guardrails in place. 

A good FDE can provide a much higher probability of successful implementations, Gold said.

Many software engineers have worried that AI would make their careers irrelevant. But the FDE role embodies where the role is going, analysts and IT experts said.

Code can now be written using human language, allowing software engineers to focus  more on outcomes than servicing code, said Alex Spinelli, senior vice president for AI and developer platforms at Arm.

“I think that’s where engineering is moving to…, much more blending of the sort of technical product management thinking, design thinking, and architecture thinking,” Spinelli said.

While AI can make engineering invisible, it also opens a toolbox to solve business problems, said Stephen Jones, CUDA architect at Nvidia. “You have more tools than you ever had before to solve problems that were previously completely unsolvable,” he said.

The FDE roll in the future might well entail reducing AI costs for customers, said Deepak Seth, senior director analyst at Gartner. “Some companies are moving towards outcome-based pricing…. [And] when people start realizing the real cost of tokens, then companies will start looking at token efficiency.”

Gold said the FDEs’ implementation efforts can help drive those token savings. “If the implementation is optimized, it can save on token costs for processing the workflows, …especially as companies move to agents,” he said.