Claude Code branding on a removable poster at Anthropic PBC’s Code with Claude developer conference in London, Tuesday, May 19, 2026.
Bloomberg | Bloomberg | Getty Images
Big Tech companies, which have spent billions of dollars developing artificial intelligence, are spending millions to build the workforce needed to prove that their investments are paying off.
Companies are scrambling to find specialized AI engineers who can leverage this technology within their companies. These professionals, known as Forward Deployment Engineers (FDEs), combine advanced AI and software expertise with an understanding of how companies operate. They can play the critical role needed to turn powerful, large-scale language models into enterprise systems that can deliver measurable business outcomes.
A number of AI and memory stocks, from Dell to Coreweave to AMD to Micron, have suffered significant short-term losses this week as market concerns that the AI boom economy is unsustainable have grown.
Amazon Web Services has committed $1 billion to forward deployment engineering initiatives. OpenAI launched its implementation business with over $4 billion in initial investment support. Anthropic is committing $100 million to a program to train 10,000 forward-deployed engineers by the end of 2027. Accenture and Google Cloud are also building a workforce of 1,000 forward-deployed engineers.
“People are becoming impatient with billions of companies investing in AI without seeing any business results,” said Jeff Christian, founder of executive search firm Christian & Timbers.
FDE works directly with customers to overcome existing technology systems, security requirements, and organizational hurdles to turn promising AI experiments into real-world applications that enterprises can use.
Val Henderson, CEO of Caylent, an AWS consulting partner that embeds FDE for its customers, said the role requires more than technical expertise.
“The most difficult thing to teach is definitely customer emotional intelligence,” she says, adding that engineers need to understand the customer and the business, navigate organizational dynamics, and make decisions when a project encounters unexpected problems.
The number of job openings has increased rapidly and salaries are paid despite a shortage of human resources.
Experienced talent is in short supply, and developing that expertise will not happen overnight in a competitive labor market that shows recent acceleration in hiring numbers.
The number of FDE jobs in the U.S. jumped from less than 200 in 2023 to about 1,200 in 2025 and more than 7,300 in the first nine months of 2026, according to data from labor market analysis firm Lightcast.
The median posted salary during this period reached $195,456, an increase of 40% from 2023 and more than three times the median posted salary for all U.S. jobs.
The competition extends far beyond the companies developing AI models. Consulting firms are among the largest recruitment firms, reflecting the demand for engineers who can help enterprise customers implement technology.
Lightcast counted 872 unique FDE jobs from Deloitte and 688 unique FDE jobs from Accenture in the first nine months of 2026. This is followed by Google with 499 and KPMG with 447. This number represents job openings, not completed positions.
“The demand for forward-deployed engineers is rapidly increasing, and rising wages suggest that employers are competing for a limited talent pool,” said Elena Magrini, global head of research at Lightcast.
But the data also points to potential solutions.
“FDE is not an entirely new role; approximately half of the skills in FDE are shared with software engineering, creating an opportunity to tap into a broader talent pool,” Magrini said.
Technology companies and their enterprise customers are investing in a variety of ways to develop deployment expertise.
AWS is building its own FDE organization, hiring experienced engineers within Amazon and AWS, as well as external talent. According to Asa Kalavade, vice president of forward deployment engineering at AWS, the company will start with a few hundred engineers.
“We are currently investing $1 billion because customers need practical help to quickly turn their AI investments into results,” Kalavade said.
On the back of an initial investment of $4 billion and an agreement to acquire Tomoro, an applied AI consulting and engineering firm, earlier this year, OpenAI is pursuing a broader enterprise deployment strategy and is expected to bring approximately 150 experienced FDEs and deployment specialists to the operation.
The AI skills gap is a “market problem”
Anthropic takes a different approach by training engineers already employed by its customers and partners. The company aims to train 10,000 FDEs by the end of 2027 through the Claude Frontier Academy. The initial resident group includes engineers from Accenture, Bain, Capgemini, Commonwealth Bank of Australia, Deloitte, McKinsey, Morgan Stanley and Novo Nordisk.
“The AI skills gap impacts all of our customers and all of our partners, so this is a market issue and Anthropic is committed to closing that gap,” said Steve Corfield, global head of business development and partnerships at Anthropic.
Meanwhile, Accenture and Google Cloud are building an FDE workforce of 1,000, relying in part on retraining experienced software engineers for customer-facing AI deployment work.
“The reality is that you can’t simply hire experienced FDEs at the scale your clients need,” said Ram Ramalingam, FDE and Software and Platform Engineering Lead at Accenture. Ramalingam said the company’s FDE training and onboarding can take 30 to 90 days, depending on the employee’s background. He added that the participants already have sufficient experience in software engineering. This program is designed not to train beginners, but to add specialized introductory features.
AWS also sees opportunities to develop expertise within customer organizations. Kalavade said many companies already employ talented engineers, but need help learning how to move AI projects into production. AWS FDE works in conjunction with these teams, ultimately enabling customers to maintain and scale their systems themselves.
Caulfield said the challenge goes beyond technical expertise. “The skills that are in short supply are engineers who understand how businesses work and can build systems that change businesses,” he said.
Anthropic believes that some of its expertise can be developed through structured engineering. The company’s Frontier Deployed Engineer Residency is inspired by medical education, pairing experienced engineers with practitioners and evaluating engineers in realistic corporate deployments.
“We modeled our training on how medicine trains physicians,” Caulfield said. According to Anthropic, participants will learn from working engineers, work on real-world cases, and undergo evaluations before practicing at their own companies. You will then receive 12 weeks of mentorship from Anthropic engineers while working on the Claude Project. The first full credentials are expected to be published in early 2027.
Too early to analyze implementation results
According to Christian, experienced FDEs can help grow the workforce by leading teams of less experienced engineers. For example, he said, someone who has completed 10 to 15 successful implementations may have the judgment to oversee a team of five to six people and pass on the expertise gained from working on complex customer projects.
Christian sees this mentorship model as a way to grow his deployment team without needing an elite engineer in every position.
But Henderson said teaching the customer-facing decisions required for complex AI deployments is particularly difficult. While technical skills can be improved through training, engineers also need experience dealing with real-world customer challenges.
“There’s no algorithm that compresses experience,” Henderson says.
For Christians, the challenge comes back to pressure from investors to produce business results. He argued that companies need engineers who can work directly with customers to translate increasingly powerful AI systems into practical applications. “The only way to make AI work is through human-to-human interaction,” he says.
The ultimate test will be whether this expanded workforce allows companies to generate a return on investment from AI that is worth the initial cost. To date, the evidence available from the FDE project is inconclusive. Case studies reported by the company show that some early adopters have seen productivity gains, but no financial benefits.
At mortgage lender PennyMac, Kayrent’s FDE worked with product managers and engineers on an AI-enabled mortgage servicing platform. According to case studies provided by AWS, some engineering tasks were completed as much as six times faster, depending on the task.
Another of Keyrent’s efforts involved Smarsh, a communications compliance software provider. The company’s case study estimates that the project will free up $5 million to $10 million in engineering and product capacity that can be redirected to other work. Kyrent confirmed that FDE participated in the project, but this figure does not isolate their contribution or represent equivalent cash savings.
Improvements in engineering productivity and capabilities are not independently verified returns on AI investments.
Anthropic’s ambitious training plan has yet to show tangible results. Caulfield acknowledged that it is still too early to have outcome data, but said he plans to track how quickly participants’ projects reach production and how widely their systems are used.
