Why the forward-deployed engineer is tech’s hottest job

Your LinkedIn is suddenly filled with posts and job ads for the forward-deployed engineer (FDE) role. What does that even mean?

The term FDE was coined by Palantir, which provides data integration and analysis platforms for governments, including the defense industry. In the military, a forward-deployed soldier is stationed overseas, ready for rapid response. In tech, this engineer is deployed to a client, or to internal developer customers, to help them prepare for and respond to complexity — like the massive change of AI.

The FDE sits somewhere between a back-office software engineer and a customer-facing software architect. It’s a job not dissimilar from a field CTO. FDEs usually play a consulting role to help customers adopt and configure existing tools or, especially with AI, to set up totally nascent tools and processes, including data preparation. And just like a deployed soldier, these engineers are working toward clear and concise “missions” tied to specific — in this case, business — outcomes.

“FDEs remain hands-on throughout the customer life cycle,” writes Prasad Rao, principal solutions architect at AWS, in a January LinkedIn post. “They don’t just design solutions or oversee delivery. They get involved in daily operations, fix problems as they arise, and adjust systems based on what actually happens in the field.”

What goes into success for this hottest job in tech? Is it just to help companies get over this next AI-generated hump? Is it all hype? Or could a forward-deployed engineer be just what your company needs to hit its business goals?

What skills does an FDE need?

The FDE sits somewhere between customer success and engineering. Like a technical consultant, an FDE usually brings full-stack technical experience and an outsider’s ability to examine complex systems from a higher level — and then to dive down into the technical solution to a business problem.

Abby Bangser applies her six years as a consultant at ThoughtWorks to her current role as founding principal engineer at platform startup Syntasso.

“As a consultant, I learned to build a high-value toolbox of skills and experiences that can be applied to many different contexts,” Bangser tells The New Stack. “This has turned out to be the most valuable lesson, as it allowed me to take the centralized patterns from our platform and help roll them out to specific teams more effectively.”

She adds, “By actively joining the teams for a few weeks, I was able to pick up work tickets with their team and share in the design of how to implement ideal patterns into their real-world and messy environment. It is that combination of confident design principles and implementation goals paired with the scrappy real-world tinkering that makes this type of role both so much fun and also so high leverage.”

The FDE must be a good listener. It’s a role that relies on tight user feedback loops.

Teams that focus on the user have a 40% higher organizational performance, according to Google Cloud‘s DORA researchers. Closeness to the user is at the core of what an FDE does. To achieve this user-centric focus, the DORA team outlines four requirements:

• Integrate user feedback loops.
• Make user metrics visible.
• Involve engineering in user research.
• Leverage specification-driven development, documenting user needs before you build.

“Building a user-centric focus,” the DORA team writes, “requires a cultural shift that moves user stories, feedback, and analytics from the periphery to the core of the development process.”

Like pretty much any job in this economy, an FDE also has to be able to tie their work to measurable business outcomes. Success in this role is still measured by the traditional metrics — quality and throughput — but also via continuous qualitative customer feedback. And, of course, adoption matters:

• Are your target users actually using what you’ve built?
• Are they using it the way you expected them to?
• Is what you built driving retention or other business metrics?

How do you beef up these social skills?

If you were sold on a career in tech because you’d prefer not to talk to people and just write code, the transition to an FDE might be a bit of a shock.

While the term has been around for more than a decade, the FDE’s sudden jump in popularity is due to a demand for more human skills to pair with technical skills, especially in the Age of AI. So what are some ways to get more social?

Sit in on some customer support calls, advises Shivdhwaj Pandey, engineering manager at Docyt, in a January post on LinkedIn. Doing so helps to surface customer pain points and real needs. Support teams can help you learn to be a great listener.

He then checks in with the support rep after each call: “Here’s what I heard. What do you think?” This helps you verify you are really hearing your customers, as well as helping you understand if it’s a one-off complaint or a pattern that needs attention.

“I used to think my job was building what Product asked for,” Pandey writes. “Now I know my job is understanding the gap between what we build and what users need.”

One long-time FDE at Palantir writes on the company blog that his job has him consistently asking the following questions:

• What products are we deploying for this use case?
• Why are we deploying them?
• How will we spin up workflows that utilize these products to address the customer’s specific needs?

Then you need to implement your potential solutions in collaboration with your end users.

Proofs of concepts (PoCs) are a great way to throw the proverbial spaghetti at the wall to see if your customer wants it to stick.

“People don’t know what they want until they see something they don’t want,” Mark Coleman, co-founder and chief product strategy officer at NetBox Labs, tells The New Stack. “Being able to get software in front of them and say: ‘Do you mean this?’ And then have them go: ‘Yeah, kind of, but, it’d actually be better if it worked this way,’ or ‘It needs something extra.’”

AI also makes this type of prototyping faster to give a sense of what you want to build and to check if your users actually want it.

Engineering leadership should support FDEs, Coleman says, by removing barriers between them and clients, so developers “can create prototypes and not have to wait for permission.”

Another FDE aptitude to work on, he adds, is building empathy.

At NetLabs, engineering teams create working groups with external partners and customers to solve problems that Coleman describes as “quite gnarly,” like one that right now is working on dense wavelength division multiplexing.

“We’re bringing people together from all sorts of companies to talk about, not only: How should we solve this problem? But also, to learn what is the state of the art in this space, and what do people imagine might come up next?”

Finally, Coleman argues that, in order to be a good FDE, you have to work on your writing skills.

“Being good at writing is a more important skill than ever now because, even though AI is able to throw out all sorts of stuff, it is still a garbage-in, garbage-out approach,” he says. “And if you’re unable to express your ideas in writing accurately, then you’re going to have less success with these technologies,” and less ability to parrot back and clarify what your users want.

He adds, “There are many areas where we’re going to find software engineers that can really lean on skills that maybe have been traditionally valued more in other areas that are going to make them more successful.”

Does AI require an FDE?

Off-the-shelf AI models and tools usually require humans in the loop to help tailor AI into your existing systems.

But while a traditional FDE is typically customizing best-of-breed, proven tools, an FDE for AI (AI FDE) lives in “ambiguity by default,” writes AI career coach and adviser Sundeep Teki. Instead of platforms, the FDE is, at least for now, focused on fine-tuning of models, large language model (LLM) deployment, and Retrieval-Augmented Generation (RAG) systems.

However, every organization’s data is like a snowflake. As MIT NANDA found in 2025, 95% of AI pilots fail in the face of enterprise silos, blocking essential data integration and experimentation lessons.

According to Thorsten Walther, managing director of the CxO advisory at MongoDB, “Data, not models, will define the next wave of enterprise AI.

“The AI story in 2026 won’t be about which large language model is ‘best,’” he tells The New Stack. “It will be the realization that models are becoming interchangeable, and the real competitive edge lies in the data that fuels them.

“Enterprises will prioritize LLM portability so they can choose the right model for the right job, and they’ll invest even more in vector search, embeddings, and re-ranking to extract deeper value from their own data. Conversations across the market are already shifting in this direction, signaling a maturing view of what actually moves the needle in enterprise AI.”

An AI FDE has to work with different departments to onboard and optimize based on different datasets, meaning they also need a solid data engineering background.

From the chief AI officer (CAIO) on down, a myriad of roles are cropping up to create an AI strategy, grounded in data and customization that supports your organization’s unique value proposition. The FDE brings both technical prowess and sales skills and then goes out to find what these new AI users — including internal users— need, and then helps build it.

The case for the forward-deployed platform engineer

AI or not, what could be a better application of the FDE role than to an internal platform team?

If you are truly going to treat your internal developers like your customers — as is the way of platform engineering — then forward deploying a platform engineer or two and embedding them in an application team is the best way to build a PoC and get truly rapid — and likely candid — feedback from your colleagues. This builds trust with your target audience that you are building based on their lived experience.

As Director of Platform Engineering, Nick Ness says his and other platform teams at 84.51˚, a retail data science, insights, and media company, embed on application teams to help jumpstart “major migrations, or to test new patterns in beta teams where we are still figuring out what changes will be made or problems that we will encounter in the real world.”

Deploying platform engineers to app teams shows a long-term commitment to your internal developer platform.

“When we talk about Platform as a Product, we often talk about needing the go-to-market skills such as [developer relations] to connect with users and make sure their questions, concerns, and needs are addressed in the product,” says Syntasso’s Bangser.

“Bringing in an FDE just takes this to the next level by investing in on-the-ground collaborations that go past whiteboarding and a proof of concept to get into the real-world challenges of onboarding, scaling, and securing the platform solutions.”

Embedding a potential FDE within your own enterprise setting is another way to let an engineer test out this more sales-oriented role before making a move into the field.

A short-term role with long-term skills

At a time when the more human side of tech is more important than ever, à la Honeycomb CTO Charity Majors’ engineer/manager pendulum, FDE can allow developers to dip their toes into more customer-facing roles.

Just allow them to move back if it’s not a good fit or if they want to take what they learned and apply it to a lead engineer or platform engineer role.

The FDE role is suddenly very popular at a time when the tech industry is in a constant state of flux. The FDE’s job description will likely change, even from week to week. Or the role may disappear in a year or two — à la the prompt engineer — as AI tooling settles and organizational data silos break down.

What won’t likely change is the increasing importance of engineers connecting what they are building to business goals and user feedback. Those are the skills developers will need in the long haul.

The post Why the forward-deployed engineer is tech’s hottest job appeared first on The New Stack.