The Forward Deployed Engineer

Forward Deployed Engineering

If you live around Silicon Valley or the software world, you're starting to hear the term Forward Deployed Engineer (FDE) more and more. While the term has its roots in the military, it was popularized by Palantir over the last 20 years. The primary distinction between a Forward Deployed Engineer and a more traditional software engineer is their relationship with customers. The latter is much more typically back-office, often purposely hidden away because of worry that they're too technical or awkward for client interactions. Meanwhile, Forward Deployed Engineers are meant to sit within organizations, actually putting the software to work.

Nic Prettejohn, head of AI in the UK at Palantir, put it well in a recent FT piece: "[Forward Deployed Engineers] know that the only valuable software is not how exquisite its code is or how beautiful the language... It's only valuable if it means something for the end customer."

Gone are the days when you have four layers of "suits" sitting between engineers and customers. Those layers were sold as quality control and strategic oversight, but more often, they actually degraded quality while adding cost and delay. Each layer introduced translation errors, approval delays, political friction, context loss, and accountability diffusion. The builder never got to see the actual problem firsthand, and the customer never met the person building the solution.

Forward Deployed Engineers collapse that distance. They sit with the customer, see the problem firsthand, and build the solution directly. Quality goes up, cost comes down, delivery speeds up. That comes less from AI writing code faster, though it does, than from removing the friction that used to degrade all three.

Palantir's Shyam Sankar has this great framing: there are engineers who know how to build the right thing (MacGyver types, whose high is "I solved the problem") and engineers who know how to build it the right way (artists, whose high is "don't you see how beautiful my architecture is?"). The best FDEs skew MacGyver. They care about whether it works, not whether it's elegant.

With AI the value comes from iteration, not the upfront plan. Each prompt refined and edge case handled makes the system meaningfully smarter, and that learning only happens from the inside.

As Noah Brier, co-founder of Alephic, puts it: "The best AI systems aren't built by people who understand AI. They're built by people who understand AI and your business. Forward Deployed Engineering is how you get both."

The behavior test (not the title test)

The fastest way to tell an FDE role from a not-FDE role is to ignore the title and answer four questions:

1. Does the work happen primarily in the customer's environment? Customer data, customer tools, customer Slack, customer office. Not the vendor's office, not a demo environment.
2. Does the person ship production code that the customer depends on? Not slideware, not a Notion doc, not a Figma. A pull request, a deployment, a running workflow.
3. Does the customer treat them, not the vendor, as accountable for the outcome? When the AI hallucinates in production, who's on the bridge call? If the answer is "the FDE," they're an FDE.
4. Does the role span both shaping the product and shipping the integration? The best FDEs feed customer patterns back into the roadmap and write the integration code at the same customer.

Four yeses and you're looking at an FDE, regardless of what HR calls them. Three yeses with a "sometimes" on the fourth is the borderline territory where Solutions Engineers, Applied AI Engineers, and Customer Engineers overlap. Two or fewer and it's not an FDE, even if the business card says so.

What FDEs build

Non-exhaustive, drawn from real listings across the industry:

• Eval harnesses specific to a customer's data and definitions of correctness
• Custom retrieval pipelines when the customer's documents don't fit a default RAG setup
• Workflow orchestration that wraps a model in the customer's approval, audit, and exception-handling logic
• Agent tooling: the function-calling layer that lets the AI take actions in the customer's stack
• Connector code that bridges the AI product into the customer's data warehouse, ticketing system, CRM, EHR, or trading platform
• Internal product feedback loops: patterns from 10 customer deployments become the next quarter's roadmap

Where FDEs come from

• Full-stack engineers at product companies who've wanted more direct customer contact.
• Consultants (Palantir alums, McKinsey Digital, BCG Gamma) who learned to code and got tired of slideware.
• Founding engineers at startups that didn't make it. They have product instincts, customer empathy, and shipping experience.
• Solutions Engineers who learned to write production code. The most direct lineage.
• Domain experts who learned to code: bankers at Hebbia, lawyers at Harvey, clinicians at Abridge. This is the newest and fastest-growing path.

What they share: comfort with ambiguity, willingness to fly, and equal fluency in code review and customer call.

FDE vs. neighboring roles

Companies hiring FDEs today

The demand is real and accelerating. Postings for the Forward Deployed Engineer title rose more than 800% between January and September 2025. By spring 2026, Google, Amazon, and Meta had each publicly committed to hiring hundreds into new Forward Deployed Engineering functions.

The role exists at scale across at least four overlapping clusters:

• Frontier model labs: Anthropic, OpenAI, Google, Mistral, Cohere
• Enterprise AI platforms: Scale AI, Databricks, Glean, Cresta, Intercom, Hebbia, Baseten
• AI-native verticals: Harvey (legal), Hippocratic & Abridge (healthcare), Decagon & Sierra & Maven AGI (customer support agents), Hebbia (finance), ElevenLabs (voice)
• AI-augmented hardware/defense: Palantir, Anduril, Gecko Robotics

The full list is on the live board, refreshed every 6 hours from the source ATSes.

Examples of FDE and FDE adjacent roles on our job board

The industry hasn't standardized, but that matters less than it looks: the function is bigger than any one company's vocabulary. Inside a single company you'll often find multiple titles doing variants of the same job. A non-exhaustive map:

A title-based taxonomy misses roles that fit and includes ones that don't.

Hire one

Every company needs a slightly different FDE. Tell us about yours and we'll generate a tailored JD: comp band, scope, day-in-the-life, interview rubric.