Forward-deployed engineer (FDE) is a role pattern that originated at Palantir and has spread across enterprise AI and infrastructure companies — Anthropic, OpenAI, Databricks, Scale AI, and most large-deal AI startups now hire under this title. The role sits between engineering and the customer: an FDE works embedded with a single enterprise account, builds custom integrations and workflows on top of the company's product, and is measured by how much value that customer extracts from the platform.
What the work actually splits into
The defining feature of the role is that an FDE is embedded with a customer for weeks or months, not a meeting at a time. Inside that embedding, three layers of work coexist.
The first is technical implementation: writing code that connects the company's product to the customer's systems, building custom UI on top of the platform, designing data pipelines that ingest the customer's data into the company's tooling. This is engineering work, not configuration work — most FDE roles include real Python, TypeScript, and SQL development.
The second is product feedback. FDEs are typically the first people inside a company who see how customers actually use the product. They translate customer pain into specifications, file detailed bug reports, and often ship the first version of features that later become product before the core engineering team productionises them.
The third is account growth. An FDE who builds something the customer values increases the contract value of the account at renewal. That is one reason the role has spread so quickly — it produces measurable revenue impact in a way that pure engineering roles do not.
The employer landscape
Enterprise AI companies are the largest single hiring vertical for FDE roles in 2025–26. Anthropic, OpenAI, Cohere, and Mistral all hire forward-deployed engineers to support their largest enterprise contracts — typically deals in the seven-to-eight-figure range where the customer needs custom integration and workflow tooling. Compensation is at or above standard senior-engineer levels; travel expectations vary widely by employer.
Data and analytics platforms — Databricks, Snowflake, Confluent — hire FDEs to support strategic accounts. The work skews toward data-pipeline implementation and custom analytics rather than ML application.
Defence and government-adjacent companies — Palantir, Anduril, Scale — pioneered the role and continue to define its most demanding form, with both clearance and travel requirements that limit who can apply.
Series B+ infrastructure startups increasingly hire FDEs as their first non-engineering customer-facing technical hires, replacing the model of dedicating engineering time to single customers. This is where most fully-remote FDE roles live; the work tends to skew lighter on travel.
What skills actually differentiate candidates
Strong FDEs combine three things that are independently uncommon and rarely combined. They write production-quality code in at least one language the customer's stack uses (Python and TypeScript cover most cases). They communicate technical decisions clearly to both engineers and non-engineers, including in writing. And they have the temperament for ambiguity — most FDE engagements begin with a customer goal that the FDE has to translate into a concrete technical plan.
The skill most often missing in candidates pivoting into FDE roles is the temperament. The role is part engineer, part consultant, part diplomat. Engineers who need a tightly-scoped backlog rarely thrive; engineers who enjoy walking into ambiguity and producing structure usually do.
A second under-appreciated skill is writing. FDEs produce a steady stream of artefacts — kickoff briefs, weekly updates, escalation memos, architecture docs — that are read by senior people on both sides. Strong written communication compresses the time required for everyone else to make decisions.
Five things worth checking before you apply
Travel expectations. Most FDE positions involve travel to customer sites, sometimes 30–50% of the time, particularly in the first months of an engagement. Get the percentage in writing before accepting.
Customer mix. Ask which customers the role would support. Working with a single household-name customer is very different from rotating across five mid-market accounts; both are valid, but the brief is different.
Reporting line. Is the FDE in engineering, in customer success, or in a dedicated FDE/professional-services org? The reporting line determines whether your work is measured in code shipped or in customer outcomes — both are reasonable but they're not the same job.
Code-vs-deck ratio. Some "FDE" roles are 80% slide decks. Read the job description carefully and ask the hiring manager what the role's most recent week looked like. If the answer is heavy on stakeholder management and light on code, the title is being used loosely.
Compensation structure. Ask whether there's a commission or expansion bonus tied to account performance. The presence — or absence — of this changes the role's incentive shape substantially.
The bottleneck at each level
Mid-level engineers rarely get hired into FDE roles. The bottleneck is judgment: an FDE on a strategic account is operating without close supervision and is often the most senior technical person in the customer-facing room. The role is hard to grow into without prior senior experience.
Senior FDEs thrive when they can work without scaffolding. The bottleneck for them is the temptation to over-engineer for problems the customer hasn't articulated yet. Strong senior FDEs ship the simplest thing that delivers value, then iterate.
Staff and principal FDEs usually own the strategic accounts — multi-year, multi-million-dollar relationships. Their bottleneck is balancing depth on a single customer against breadth across the FDE org's evolving practice. Best ones double as a player-coach, mentoring less-senior FDEs through difficult engagements.
Across all levels, the operating bottleneck is the customer's organisational complexity, not the FDE's technical depth. Engineers who treat customer politics as someone else's problem rarely thrive; engineers who treat the customer's internal dynamics as a system to understand usually do.
Pay and level expectations
FDE compensation usually tracks senior-engineer compensation at the same company, sometimes with a customer-success-style bonus or commission tied to account expansion. Cash compensation in 2025–26 typically falls in the $180k–$280k range for US-based roles at well-funded AI and infrastructure companies, with equity grants comparable to senior-engineer offers.
Compensation can be substantially higher at companies where FDEs are explicitly tied to revenue. Palantir-style commission structures can push total compensation into the $400k+ range for top performers. Compensation tends to be lower at companies where the role is treated as a customer-success function with engineering work attached.
European compensation typically lands in the €110k–€180k range for senior FDE roles, narrowing the gap on fully-remote roles at US-headquartered AI companies.
Equity is usually equivalent to senior-engineer grants at the same level. The expansion bonus, when present, is an annual or quarterly payout, not equity.
What the hiring process looks like
The process usually has four to five stages over three to five weeks: a recruiter screen, a technical interview, a customer-scenario role-play, an executive conversation, and references.
The technical interview rarely involves leetcode. It typically involves walking through a system you've built before, discussing trade-offs and what you'd change. Strong candidates demonstrate that they can work the abstraction stack — from API design down to deployment topology — fluently.
The customer-scenario stage is the highest-signal one. You are given a brief — sometimes a redacted real customer engagement — and asked how you would scope it, what you would build first, and how you would handle the most likely failure modes. This is the closest the loop comes to simulating the actual job.
The executive conversation is usually with a VP-level sponsor of the FDE function. The questions are about judgment under ambiguity, comfort with senior-customer interactions, and how you handle situations where engineering and the customer disagree.
References are bidirectional. The hiring company will check yours. You should check the team's by talking to two or three current FDEs about what their last six months actually looked like.
Red flags and green flags
Red flags. Vague answers about travel expectations. The role's recent week is described in stakeholder-management terms with no code shipped. The reporting line and incentive structure are unclear. The company describes the role as "doing whatever the customer asks" — a legitimate FDE function has clear boundaries and a defined practice manager. The hiring manager cannot name the FDEs you would be peers with. The product itself is poorly documented internally — FDE work compounds when the underlying product is well-documented.
Green flags. A clear practice owner who runs the FDE function as a discipline. A defined onboarding programme — the strongest FDE orgs run a four-to-six-week ramp. Public engineering blog posts written by current FDEs. A clear staffing model that rotates FDEs off accounts at predictable intervals to avoid burnout. Travel expectations stated in the job description or in the recruiter screen, not buried in the offer letter. Compensation structure that's transparent on commission/expansion components.
Gateway to current listings
Below are remote forward-deployed engineer roles currently active in the RemNavi corpus, sourced from the major remote job boards and direct ATS feeds. Listings refresh daily.
Frequently asked questions
What is the difference between a forward-deployed engineer and a solutions engineer?
Solutions engineers operate pre-sale; their job is to help close deals. Forward-deployed engineers operate post-sale, embedded with a customer for weeks or months, and their job is to make the customer succeed with the product. Compensation, incentive structure, and time horizon are all different.
Do FDE roles require travel?
Most do, in the 20–50% range. Some are fully remote. The travel expectation is the most important question to ask when evaluating a role and is often under-disclosed in job descriptions.
Is FDE a path into product management or engineering management?
It can be either. FDEs see how products are used in the wild, which is excellent preparation for product management. They also work at the boundary between engineering and customer needs, which sets up well for engineering management roles where customer judgement matters. The path is less common than the engineering→PM or engineering→EM paths but is well-trodden at companies like Palantir.
What programming languages should an FDE know?
The right answer depends on the customer's stack, but Python and TypeScript cover the majority of cases. SQL is essential. Most FDE roles also benefit from comfort with at least one cloud platform's primitives — typically AWS, GCP, or Azure depending on the employer's customer base.
How do FDE compensation packages compare to standard senior-engineer roles?
At most companies they track senior-engineer compensation closely, sometimes with a small commission or bonus component tied to account expansion. At companies where FDEs are explicitly revenue-attached, total compensation can exceed standard engineering roles substantially.
Related resources
- Remote solutions engineer jobs — the pre-sales counterpart
- Remote sales engineer jobs — adjacent role with stronger commercial focus
- Remote applied ML engineer jobs — common FDE specialisation in AI companies
- Remote technical account manager jobs — non-coding peer role
- Remote founding engineer jobs — adjacent high-ambiguity engineering role
- Remote developer advocate jobs — another customer-facing technical role