FDE Postings Jumped 729%. The Title Hides Five Different Jobs.

If you've opened an Indeed search for "Forward Deployed Engineer" in the last 30 days, you've seen the chaos. The same three-word title is attached to a $53,000 implementation gig in Charlotte and a $1.2M comp band in Mountain View. Recruiters who treat that as one funnel are burning weeks on the wrong pool.

April 2026 Indeed data shows FDE postings went from 643 to 5,330 year over year, a 729% jump. FDE Pulse's April 30 snapshot of 247 active postings shows Google (34) and Deloitte (19) together own 40% of the visible market, while 131 other companies posted only one to three roles each. That distribution alone tells you the title isn't standardized. It's being arbitraged.

The 729% number is real. The job underneath it is five jobs.

A16z's framing of "job-title arbitrage" was originally about Palantir using FDE to signal a new capability. In 2026, consultancies and SaaS vendors are re-arbitraging the same words to make staff augmentation and solutions engineering sound prestigious. The result is one keyword covering five distinct hiring markets.

729%

Here's the breakdown anyone doing forward deployed engineer hiring needs to internalize before they write a single boolean string.

1. AI lab FDEs ($350K to $550K TC)

OpenAI and Anthropic have stabilized FDE comp at $350,000 to $550,000 for mid-to-senior levels. Both labs announced billion-dollar FDE ventures within days of each other in May 2026. OpenAI's acquisition of Tomoro, an applied AI consulting firm, brought roughly 150 engineers with prior deployment experience at Tesco, Virgin Atlantic, and Supercell. These are not solutions engineers. They report into product or research, ship code into customer environments, and the bar is closer to a senior product engineer at a top lab than a field consultant.

2. Big Tech cloud FDEs (~$238K average)

Google Cloud's "Senior Staff Forward Deployed Developer" tops the disclosed market at $262K to $365K base. Thomas Kurian has publicly confirmed accelerated FDE hiring at Google Cloud. KPMG, ServiceNow, and Vista Equity Partners also list roles above $300K max. These roles look like a hybrid of TAM, customer engineer, and pre-sales architect with deeper code expectations than the old "sales engineer" mold.

3. Big Four and consulting FDEs ($150K to $300K base, billable)

Deloitte's Charlotte listing demands 10 to 15+ years in software engineering or platform delivery. PwC posts Director-level "Forward Deployed Software Engineer, Palantir Foundry" roles. McKinsey QuantumBlack's "Lead Forward-Deployed Engineer" requires more than eight years of hands-on platform or infrastructure work plus a technical degree. Collectively, Deloitte, Accenture, KPMG, BCG, PwC, and McKinsey post more FDE roles than any product company except Google. If you find an "AI FDE" on Indeed, the prior is now that it's a Foundry or agentic-AI consultant, not a lab engineer.

4. Series A to C AI-native startup FDEs ($180K to $250K plus meaningful equity)

Modal Labs, Cursor, ElevenLabs, Scale AI. These shops are sourcing heavily from Palantir alumni and from each other. Cash bands look uncompetitive on paper because 0 of 196 active FDE postings mention equity. The published numbers systematically understate real TC at every startup in this tier.

5. Implementation and field engineering mislabeled as FDE ($53K to $120K)

This is where the bottom of the $53K to $1.2M spread lives. Field service, deployment, or implementation roles relabeled with the trendier title. They are not bad jobs. They are just not the same job, and conflating them is the single most common FDE recruiting mistake of 2026.

The reporting line is the cleanest signal

A useful test, faster than reading the JD: who does the role report to? If it reports into engineering or research, you're looking at a true FDE. If it reports into sales or customer success, it's solutions engineering wearing a more expensive jacket. The Aaron Levie line going around, that FDEs are "one of the most in-demand roles in tech, and one of the most critical functions for bringing AI to life," is true for the first interpretation and largely marketing for the second.

The reporting line tells you more than the JD does. Engineering means FDE. Sales means SE with a wig on.

This matters because the candidate pool you draw from is completely different across the five tiers. A Palantir FDSE alum will close fast on an Anthropic role and ghost on a Deloitte Foundry req. A senior Accenture consultant looks great on paper for the Deloitte role and will not survive an OpenAI loop. The same boolean returns both candidates. That's why generic title searches across LinkedIn and Indeed produce shortlists that interviewers reject in week one, and it's why we built Refolk: you describe the actual person you want in plain English, including the reporting line and the prior employer pattern, and get a ranked shortlist instead of a title match.

Supply is roughly 3.4 jobs per qualified head

Here's the part nobody is pricing in. Refolk's index returns about 1,575 U.S. professionals using FDE-family titles. Against the 5,330 open postings, that's roughly 3.4 jobs per qualified candidate, and the supply is heavily concentrated.

1,575

New York leads the supply distribution, matching the demand shift (New York is now 35% of postings, surpassing San Francisco's 11% because of fintech and regulated-industry demand). Palantir Technologies shows up roughly 5x more frequently than any other employer in the supply pool. Salesforce, Modal Labs, Cursor, ElevenLabs, and Scale AI round out the top of the index.

If you take only one tactical thing from this piece, take this: Palantir alumni are the highest-leverage outbound target for any true FDE req, full stop. Every AI-native shop hiring forward deployed engineers in 2026 already knows this. The ones with messy sourcing pipelines are competing for the same 200 names. The ones running tighter pipelines are expanding the pool by mapping second-degree alumni networks (engineers who joined Modal or Cursor from Palantir, and the people who joined those companies after them).

The 34% remote tell

83 of 247 active FDE roles, or 34%, list as remote-eligible. The other 164 are on-site or hybrid. That's a lower remote share than the broader software engineering market, and it tracks: direct customer presence is part of the value an FDE delivers. If your pipeline skews toward fully remote candidates, you're already filtered out of two thirds of the market. Anduril's "Forward Deployed Engineer, Air Defense" role in Costa Mesa is an extreme case (defense tech with clearance and a physical site), but the principle holds across the consulting and lab tiers too.

How to source the right flavor

Once you know which of the five FDE markets you're hiring in, sourcing collapses into a tractable problem. The trick is matching the candidate's prior pattern to the actual role, not to the title on the JD.

For AI lab FDEs, pull Palantir alumni currently at Modal, Cursor, ElevenLabs, Scale AI, and lateral peers at other labs. Filter for engineers who shipped customer-facing systems, not internal infra. For Big Tech cloud FDEs, the cleanest source pool is current TAMs and customer engineers at AWS and Azure with measurable code contributions on GitHub. For Big Four FDEs, target senior Foundry-certified consultants and Palantir alumni who left for consulting (a smaller but real cohort).

For Series A to C startups, the equity disclosure gap (0% of postings mention equity) means your outbound has to do the math the JD won't. Candidates from Anthropic, OpenAI, and Palantir know what equity is worth at scale; candidates from Big Four shops often don't. The conversation is different.

For the low-end implementation roles mislabeled as FDE, the honest move is to rename the req. You will lose half your applicants and gain a shortlist of people who actually want that job.

What the salary-source disagreement is telling you

ZipRecruiter lists the average FDE salary at $116,463 as of May 2026. Glassdoor lists it at $155,594. The same title, the same time window, averages off by about $40K across two well-known data sources. That gap isn't noise. It's the definitional chaos of the title surfacing in the data. ZipRecruiter's sample skews to contract and junior implementation roles. Glassdoor catches more of the consulting and Big Tech bands. Neither captures the lab tier cleanly because equity is missing.

Anyone quoting "average FDE salary 2026" without naming the tier is doing the same thing as the JD writers: arbitraging the title. The $53K to $1.2M spread is honest. The averages aren't.

The 12-month forecast nobody is writing

If postings grew 729% in 12 months on a base of 1,575 qualified candidates, the obvious prediction is that the title fragments. Expect "Forward Deployed AI Engineer," "Forward Deployed Solutions Architect," "Forward Deployed Research Engineer," and "Forward Deployed Implementation Engineer" to appear as distinct reqs by Q3 2026, mostly driven by the consultancies that want to keep arbitraging the prestige without competing for Palantir alumni.

When that happens, the recruiters who already built mental decoder rings (and who already have a shortlist of the 1,575 real ones tagged by tier) win the next 18 months of AI deployment engineer sourcing. The ones still running blanket title searches will keep wondering why their close rates are dropping while their pipeline volume looks fine.

FAQ

What's the difference between a Forward Deployed Engineer and a Solutions Engineer in 2026?

The cleanest distinction is the reporting line. A true FDE reports into engineering or research, owns customer-deployed code, and is measured on shipped systems. A Solutions Engineer reports into sales or revenue, owns deal support and demos, and is measured on pipeline influenced. Title overlap is enormous in 2026 because consultancies and SaaS vendors have re-labeled SE roles as FDE for prestige and comp signaling. Always check the org chart before you source.

Why are Google and Deloitte hiring so many more FDEs than everyone else?

Google Cloud is using FDE as the name for its senior customer-facing engineering bench as Gemini and Vertex deployments scale. Deloitte and the other Big Four are using it as the title for their Palantir Foundry and agentic AI delivery practices. Both are real hiring needs, but they pull from totally different candidate pools (lab and Big Tech engineers vs. management consultants with platform experience). They show up as 40% of postings because both are running well-funded, centrally coordinated programs while the rest of the market is hiring one or two FDEs at a time.

How do I source Palantir alumni without spamming the same 200 names everyone else is messaging?

Go two degrees out. Map the engineers who joined Modal, Cursor, ElevenLabs, and Scale AI from Palantir, and then map who joined those companies after them. The second-degree cohort is harder to find on LinkedIn because the title transitions don't always include "FDE" or "FDSE" in the path. This is exactly the kind of query where natural-language sourcing tools like Refolk outperform boolean: you can ask for "engineers two hops out from Palantir FDSE alumni at AI infra startups" and get a shortlist instead of a title match.

What should I budget for an FDE hire in 2026?

It depends entirely on which of the five tiers you're hiring into. Lab FDEs run $350K to $550K TC heavy in equity. Big Tech cloud FDEs cluster around $238K average with Google's Senior Staff band reaching $365K base. Consulting FDEs sit at $150K to $300K base with billable utilization expectations. Series A to C startups land at $180K to $250K plus meaningful equity that won't show up in the posting. If you're seeing $53K to $120K bands labeled FDE, that's implementation work and should be sourced and titled differently.