A six-month study by an executive search firm finds that roughly 2,000 people in the United States have the right mix of hands-on AI experience, industry knowledge, and trust from business leaders to help large companies get real returns on their AI investments. The study, conducted by Christian & Timbers and shared exclusively with TechCrunch, puts it bluntly: "Not 2,000 available. 2,000 total." TechCrunch

The role these people fill is called a forward-deployed engineer, or FDE. The concept originated at the data analytics company Palantir. The idea is simple: instead of building software in isolation, these engineers work directly inside client companies. They combine technical skill with enough industry knowledge and people skills to turn AI tools into systems that actually work in the real world. The report identifies about 17,000 FDEs in the U.S. today, many already employed at Palantir, but only about 2,000 of them have what it takes to guide company-wide AI overhauls.

Jeff Christian, founder of Christian & Timbers, led the research from January through June 2026. His team interviewed more than 250 senior executives at 180 companies, surveyed 80 Fortune 500 executives, and spoke with more than 300 FDEs and applied AI engineers.

The demand has climbed fast. At the start of 2026, only 5% to 10% of companies were planning to hire FDEs, mostly for small trial projects. By the end of the second quarter, that share had jumped to 70%. The study projects demand to surge by 2,100% by the end of 2026.

The largest consulting firms said they need to multiply their FDE teams by ten, building groups of 20 to 100 people instead of hiring one or two specialists for short-term work.

The numbers point to a deeper problem. If 70% of companies now want FDEs and only 2,000 qualify, this is not a typical hiring challenge. It is more like a generation problem. The required skills take years to build: deep knowledge of a specific industry, repeated experience with real AI projects, and the kind of presence that earns trust in a boardroom. None of that comes from a short training program.

There is an important difference between the 17,000 FDEs and the 2,000-person top tier. The larger number shows the job title itself is not rare. What is rare is the combination of deep expertise and credibility needed to steer major AI investments toward measurable payoffs. A company hiring from the broader pool may get someone who can connect AI models and build data systems. What they may not get is someone who can walk into a boardroom, figure out why an AI project has stalled, and fix both the technology and the team dynamics around it.

The timing matters because companies have been pouring money into AI, but many are still waiting to see that spending pay off. If FDEs are the people who close that gap, then having so few of them limits how quickly the current wave of AI investment turns from cost into return.

Whether this gets solved through new training programs, promoting people from within, or rethinking how AI projects are staffed is an open question. The study does not propose a fix; it measures the gap. The arithmetic is harsh either way. Two thousand people, no matter how you count them, is not a hiring pipeline. It is a bottleneck.