The value of the title FDE (forward-deployed engineer) lies not in its freshness, but in how it redefines a class of work that was previously underestimated: on-site technical implementation for customers.

In traditional software companies, this work is often placed in the borderline areas of pre-sales, implementation, solution engineering, or customer success. It is very close to the customer and the product, but frequently occupies a marginal position in organizational narratives.

Palantir recognized this early on.

Around 2011, it renamed engineering roles that were originally oriented towards customer sites and system integration to FDE. Behind this naming was a clear judgment: in large enterprises and government clients, the real challenge is not writing the software, but getting the software into the client's real business systems. Permissions, data, processes, legacy systems, organizational responsibilities—all of these are involved.

Those who can accomplish this should not be simply categorized as after-sales support or project implementation.

They represent a new organizational capability.

a16z calls this approach title arbitrage, which can be understood as "job title arbitrage": when a certain capability quickly becomes important within an organization, but the old job title has not had time to reflect its value, those who name it first have the opportunity to occupy talent, power, and market mindshare.

This approach is very interesting and worth referencing, especially for AI founders engaged in B2B business.

Job titles are essentially an organizational language

Many companies underestimate the role of titles.

On the surface, a job title is just a line of text in an HR system. But within the company, it is actually a form of organizational language. It tells others what this person is responsible for, what capabilities they represent, and whether they are qualified to participate in certain decision-making processes.

Titles like CEO, CTO, and CFO are not just descriptions of division of labor; they also serve as power identifiers. The same applies to vice president of manufacturing, product head, and growth lead. What these names correspond to is the organization's recognition of certain capabilities.

This is also why job titles evolve continuously with changes in the industry.

In earlier years, those who wrote code were often classified under IT. Later, they became known as programmers, and then as software engineers. This change is not just a play on words; it reflects the rising status of software within commercial systems. Writing code transformed from backend support into a core capability for building products, organizational processes, and business models.

Data roles follow a similar trajectory. From clerk to data entry, to data scientist, and then to machine learning engineer. Each title change corresponds to the rising strategic value of data work.

The site reliability engineer proposed by Google is also a typical example. It redefined the work of traditional system administrators as engineering problems, expressing a judgment: keeping systems stable is as technically challenging as developing new features.

Therefore, a job title is not merely packaging.

It reflects whether the value of a type of work has migrated.

Palantir captured the recruitment mindset

The reason FDE became a classic case is that it transformed on-site engineering roles for customers from undervalued positions to high-potential roles.

In many companies, the status of on-site technical work is unclear. It is too close to sales, making it easy for engineering teams to consider it "not pure enough"; it is too close to delivery, leading management to see it as a cost center. As a result, truly exceptional engineering talent may not be willing to take on this role.

Palantir's renaming changed the narrative.

The message it conveyed is: you are not doing ordinary after-sales support, nor are you delivering external projects. You are solving the most complex problems on-site for customers, connecting real business systems with the company’s products.

This narrative attracts a type of hybrid talent: someone who can code and face customers; who understands systems and can handle organizational complexity; who can solve current problems and bring on-site experiences back to the product.

For such individuals, seeing titles like “implementation engineer” or “solution engineer” may suggest limited job scope. But if they see FDE, their perception will be entirely different.

This is the recruitment advantage brought by naming.

To this day, whenever FDE is mentioned, many people's first reaction is still Palantir. Not because only Palantir can do this kind of work, but because it was the first to associate this term with its company capabilities.

The one who names it first often occupies the mindset first.

The difference between new titles and false gilding

Of course, not all new job titles have value.

Some are just title inflation. For example, renaming a marketing associate to growth strategist without changing the job content; or changing an assistant to head without altering decision-making power. These types of naming can only provide short-term dignity and cannot create real talent attraction.

The original text provides a good criterion:

Would someone from five years ago find the work described by this new title unfamiliar?

If the answer is yes, then it might correspond to a new ability. For instance, the GTM engineer proposed by Clay and the legal engineer suggested by Harvey are not mere renaming of roles. They point to new combinations that emerged following AI: understanding business processes while also grasping automation; comprehending specialized contexts while being able to integrate workflows into systems.

But the prompt engineer is another example.

This term was once very popular but quickly felt outdated. The reason is that writing prompts has not stabilized as an independent profession. It resembles a foundational skill that all knowledge workers should master. If a title emerges before a real job does, its popularity will quickly wane.

Thus, the key to judging whether a new job title is valid lies not in its novelty, but in whether there is a real new job behind it.

No new work, just new packaging equals title inflation.

AI changes organizations, not just making tools smarter

The most valuable part of this article is that it places job titles in the organizational context of AI transformation.

When many companies discuss AI transformation, the default answer is: the interfaces will be smarter, tools will be more automated, and processes will be more efficient.

All of these are valid, but not sufficient.

The deeper change is that a new group of highly leveraged individuals will emerge within organizations. They may be young and originally hold lower positions, but because they know how to use AI, build workflows, and convert ambiguous problems into automated systems, they will begin to gain influence that previously did not exist.

Every time large companies introduce key software, similar phenomena occur.

The individuals who first understand the new tools are often not the highest-ranked, but the quickest to act. They are the first to recognize which processes can be restructured, which jobs can be automated, and which previously neglected issues can be reorganized.

Technology changes not just toolbars.

It will also change the distribution of power within organizations.

At this point, a new title will become important. It provides legitimacy to these individuals and offers recognition mechanisms for the organization.

For example, a legal practitioner, who was initially just interested in AI tools, starts studying contract modifications, risk control, and legal workflow automation. If the company defines this role as a legal engineer, this person is no longer just someone who "loves messing with new tools," but rather a new position that can be recognized, authorized, and promoted.

The hardest part of AI transformation is often not that employees do not know how to use tools, but that the organization lacks the language to acknowledge those already creating new value.

For AI entrepreneurs, naming is also a strategy

If you are engaged in B2B AI, the insights from this article are direct.

Don't just name the products; think about what new roles your product will create within the client organization.

If you are serving the legal industry, those early users emerging may no longer just be lawyers or traditional legal operations, but legal engineers. If you serve sales and growth teams, GTM engineers may emerge. If you serve financial research or consulting, there may also be intelligence engineers in the future.

These names are not just propagating slogans.

They will help mobilize organization internally: who should be empowered, who should be heard, and who represents this new capability.

This is also where title arbitrage brings value to companies.

Products are sold externally, while job titles spread internally within the client organization. If a new job title genuinely holds up, it will create a mindset for the product in return. In the future, when the market thinks of such roles, it will think of who proposed it earliest, who understands it best, and who can empower these individuals.

What Palantir gained from FDE is these kinds of dividends.

Back to FDE

Why is it worth discussing FDE again today?

Because the boundaries of products and services for AI-native companies are becoming increasingly blurred.

It is not always easy to distinguish whether an AI enterprise software is purely a product, a product with services, or a productized service. The process details on-site with customers can in turn define product roadmaps; model failure samples can become the capabilities in the next version; and implementation teams are no longer just at the delivery end, but are part of the product learning system.

In such cases, old titles may undervalue new capabilities.

If it is called after-sales, engineers might be reluctant to join; if it is called implementation, investors might worry about profit margins; if it is called customer success, the product team may not view it as a product signal. But if it essentially transforms complex demands into replicable capabilities on-site, then FDE is more accurate than old terms.

Of course, renaming is not a panacea.

Simply changing customer success to FDE will not automatically complete an organizational upgrade. What truly needs to change are reporting lines, incentive mechanisms, recruitment standards, product feedback systems, and how founders view the concept of "service."

The name is just the first step.

The key is whether the organization truly places these individuals at the core of product learning and customer delivery.

The emergence of a new job title often indicates that the old organizational language is no longer sufficient. Many problems facing AI companies today are precisely those that old language cannot accurately describe: products like services, services like products; engineers need to enter customer sites, and the customer site is defining the product roadmap; after-sales is no longer just a cost center, but part of a learning system.

This could be the critical watershed for the next generation of AI enterprise software companies.

It may not necessarily be who completely eliminates service; rather, it is more likely to be those who can rename, reorganize, and productize the part of service that is closest to real customer problems and most capable of forming product insights, who will build deeper barriers.

Whoever clarifies this first will plant their flag in the customer mindset.