Editor's Note: As companies like OpenAI and Anthropic begin to form AI Forward Deployed Engineer (FDE) teams, an old role originating from Palantir is regaining popularity in Silicon Valley. The core value of an FDE is to work on-site with clients, transforming general-purpose large language models into customized Agent workflows that fit specific business processes.
However, what this article truly discusses is not just the new FDE profession, but rather how job structures are being redefined in the AI era. The author believes that compared to a small number of FDEs stationed within client organizations to facilitate the implementation of specific vendor products, there will be a greater future demand for companies' own AI Engineers. They need to understand prompts, Agent frameworks, and evaluation systems, and also know how to use AI programming tools like Claude Code, Codex, etc., to truly embed AI capabilities into software and business systems.
This also implies that the impact of AI on the job market may not be a simple "replacement." It is more likely to first create a batch of new generalist roles, and then, much like how software engineers diversified into front-end, back-end, mobile, and DevOps in the past, continue to evolve into more specialized professions such as LLMOps, Evals Engineer, and AI Data Engineer. What will be truly scarce are those who understand both engineering implementation and business scenarios.
Here is the original text:
A new role has recently emerged in Silicon Valley drawing significant attention: the AI Forward Deployed Engineer (FDE). These engineers are deployed within client organizations to help customize solutions, such as building and fine-tuning Agent workflows that meet the client's specific needs. Since OpenAI and Anthropic started forming new teams to deploy FDEs into client organizations, I've heard many people refocusing on this career path.
The rise of the FDE role driven by AI workloads is an example of AI creating new jobs. It also shows that the narrative of an impending "jobpocalypse"—a collapse of the job market—is unfounded; there will still be plenty of AI and non-AI related jobs in the future. However, as explained below, I believe the number of AI Engineer positions will far exceed that of FDEs.
The FDE role was pioneered about two decades ago by Palantir. Back then, Palantir would send engineers to work on-site at government agencies, often in secure, air-gapped environments. Beyond strong technical skills, FDEs also needed communication abilities and sometimes business acumen. For example, they might need to communicate with clients to understand needs, strategize project prioritization, explain complex technology, and provide respectful yet firm feedback when clients propose unrealistic requests. The renewed focus on FDEs today is largely because truly embedding an off-the-shelf large language model into business operations and customizing it into Agent workflows tailored to specific needs requires extensive hands-on implementation work.
Nevertheless, I believe the scale of the AI Engineer role will be much larger. A company might accept a few FDEs for internal collaboration, but most companies will want more of their own employees involved in project development. At my organization, for instance, we do hire FDEs, but we hire many more AI Engineers. Additionally, a common concern among clients is the difficulty in finding truly "vendor-neutral" FDEs. After all, the FDE's mission is inherently to deeply integrate a specific vendor's product into enterprise systems. At this stage, it's hard to predict which AI service will be the best choice a year from now, so "optionality"—the ability for an enterprise to choose the most suitable vendor in the future—is crucial. In contrast, having FDEs deeply tie a company's business processes to one vendor significantly reduces this optionality.
Currently, I see market demand for AI Engineers rising rapidly. These engineers can build applications using AI software components such as LLM prompts, Agent frameworks, and evaluation systems; they can also efficiently use AI programming Agents like Claude Code, Codex, Antigravity CLI, and OpenCode. As the AI Engineer role matures, I expect it to further split into more specialized positions. This is similar to how the general "Software Engineer" role decades ago gradually diversified into front-end, back-end, mobile, data engineering, DevOps, and others.
What specialized AI engineering roles will emerge in the future? I'm not certain yet. There might be AI FDEs, LLMOps Engineers, Evaluation Engineers, AI Data Engineers, Harness Engineers, and some new roles we haven't named yet. But for now, at least, many generalist AI Engineers are already creating significant value. Excellent AI Engineers are in high demand and short supply. As this field continues to mature over the next decade, I also look forward to more specialization within AI engineering, creating further new employment opportunities.
