FDE Postings Are Up 729%. The Real Pool Is 8,821, Not 2,908.
Forward Deployed Engineer demand jumped 729% YoY and NYC passed SF. Why title search misses two-thirds of the real pool, with numbers from Refolk's index.
Forward Deployed Engineer postings on Indeed went from 643 in April 2025 to 5,330 in April 2026, a 729% jump, and somewhere in that spike the sourcing playbook broke. NYC quietly overtook SF as the primary hub. And the candidate pool most recruiters are searching, the exact-match "Forward Deployed Engineer" title, is roughly one third of the pool they should be searching. If your pipeline is empty, it isn't the market. It's your boolean.
What actually happened to FDE demand in 2026
Demand for Forward Deployed Engineers grew 729% year over year on Indeed, from 643 postings in April 2025 to 5,330 in April 2026, with NYC now holding 35% of postings versus San Francisco's 11%. Paraform's platform data corroborates the shape: FDE postings on Paraform grew 350% YoY from Q1 2025 to Q1 2026. Mentions of "FDE" in company documents are up 17% over the past six months per AlphaSense.
A Forward Deployed Engineer, first defined by Palantir in 2011, is a customer-embedded engineer who merges application and integration work: they ship code inside the customer's environment rather than selling it or supporting it from a distance. It is not a sales role. In an analysis of 1,000 FDE postings, 0% mentioned a sales quota.
The concentration is barbelled. Google (34 active postings) and Deloitte (19) account for roughly 40% of live FDE reqs. The other 60% is spread across 130+ companies each hiring 1 to 3. That distribution matters: the median hiring manager filling an FDE seat this year has never filled one before, and most of them will misfire on the JD.
Why NYC just passed SF
NYC's rise to 35% of FDE postings versus SF's 11% is a regulated-industry flip, not a general tech-industry flip. Fintech, insurance, and compliance-heavy buyers need engineers who can sit inside the customer's SOC 2 boundary and ship. That work does not happen from a Mission District coffee shop.
The mechanism is boring and structural:
- Regulated buyers require on-premise or VPC deployments, which require engineers on the customer floor.
- FDE roles require on-site embedding up to 50% of the time, so remote-only postings barely exist.
- NYC is where the fintech and gov-adjacent buyers are, so that is where the reqs cluster.
If you are running a forward deployed engineer NYC search, the target profile is not "ex-FAANG generalist willing to relocate." It is a mid-senior engineer with financial-services or gov/defense exposure who has already survived a customer-facing deploy. Weighting by AI-lab pedigree alone will underperform.
Any boolean pinned to Forward Deployed Engineer mechanically caps recall at about 33%. </pull> ## The pool is 3x bigger than your title filter shows In Refolk's index, the exact-title "Forward Deployed Engineer" pool in the US is 2,908 professionals. The sibling-title cluster (Solutions Engineer plus Deployment Strategist plus Applied AI Engineer) is 8,821. That is a 3.03x multiplier hiding in plain sight. Palantir itself splits its post-sales engineering headcount across "Forward Deployed Engineer" and "Deployment Strategist." OpenAI, Anthropic, Cresta, and Scale run similar variations. The job function is identical. The title is a marketing choice. | Title cluster | US professionals | Top employer | NYC vs SF (sample) | Source | |---|---|---|---|---| | "Forward Deployed Engineer" (exact) | 2,908 | Palantir | 7 vs 5 | Refolk's index | | Solutions Eng + Deployment Strategist + Applied AI Eng | 8,821 | Palantir | 6 vs 1 | Refolk's index | | FDE + Python skill filter | 356 | Cresta / Palantir / Basata / Northslope | 7 vs 5 | Refolk's index | | FDE postings, US (Indeed) | 5,330 (Apr 2026) | Google (34), Deloitte (19) | 35% vs 11% | Indeed / Paraform | | Sibling-to-FDE ratio | 3.03x | - | - | Derived from Refolk's index | | NYC:SF posting ratio | 3.18x | - | - | Derived from Paraform | The practical takeaway: any sourcer running a LinkedIn boolean pinned to "Forward Deployed Engineer" is capping recall at about one third of the addressable pool before they even open a message.
stat number: 8,821 label: US professionals in the FDE sibling-title cluster note: 3.03x the 2,908 who currently use the exact FDE title, per Refolk's index. </stat>
Why the Python filter kills your funnel
Only about 12.2% of US professionals with an exact "Forward Deployed Engineer" title list Python as a skill on their profile, based on Refolk's index (356 of 2,908). That is counterintuitive: Python and JavaScript proficiency is table stakes for the job. The gap is not skill, it is skill-tag hygiene.
The mechanism is identity. FDEs self-identify as customer-facing, not as ICs. They optimize their profiles for hiring managers who care about deployment stories, industry verticals, and named customers, not for keyword parsers. So the skill grid stays sparse even when the person can write CUDA in their sleep.
What to do instead:
- Filter by adjacent employer signals (Palantir, OpenAI, Anthropic, Cresta, Scale, Modal, Roboflow, Gecko Robotics) rather than skill tags.
- Filter by past-project language in the summary or headline ("shipped," "embedded," "deployed at," "customer 0").
- Ignore the skills chip grid entirely on this cluster. It is noise.
This is the exact gap Refolk closes: you describe the person in plain English ("post-sales engineers who deployed LLM products inside a Fortune 500 bank in the last 18 months") and get a ranked shortlist, instead of triangulating through three broken filters on LinkedIn.
The compensation reality: $172K to $555K depending on the logo
Median FDE total compensation runs from around $172K at the low end to $555K at the top of the market, with the spread driven almost entirely by the employer logo. Indeed's average FDE base salary sits at $171,911. Palantir FDSE median TC is about $215K. OpenAI is around $555K. Anthropic mid-senior packages sit between $350K and $550K, with public postings at $200K to $300K base.
| Company | Comp signal | Notes |
|---|---|---|
| Indeed market average | $171,911 base | Broad market, per Indeed |
| Palantir FDSE | ~$215K median TC | Coined the role in 2011 |
| Anthropic | $200K to $300K base; $350K to $550K TC | Firm on offers |
| OpenAI | ~$555K median TC | Hiring FDEs in NYC, Seattle, DC, London, Munich, Paris, Singapore, Tokyo, Sydney, Dublin |
| Google Cloud | Not disclosed | Thomas Kurian confirmed FDE ramp; 34 active postings |
| Deloitte | Not disclosed | 19 active postings; consulting-major entry point |
The Palantir FDE hiring comp anchor is worth internalizing: at $215K median TC, Palantir is well below OpenAI and Anthropic and only modestly above the Indeed market average. Founders trying to poach from Palantir have room. Founders trying to poach from OpenAI need to bring a real number.
How to hire forward deployed engineers without title search
To hire Forward Deployed Engineers in 2026, source across the sibling-title cluster, weight for regulated-industry deployment scars, and budget for 50% on-site travel. Here is the working playbook.
1. Widen the title net first
Search "Forward Deployed Engineer" OR "Deployment Strategist" OR "Solutions Engineer" OR "Applied AI Engineer" OR "Forward Deployed Software Engineer" OR "Customer Engineer." Then de-dupe by employer. The pool triples immediately.
2. Weight by deployment scar, not lab pedigree
The best FDEs have shipped inside a customer's environment against a real deadline. Signals to prioritize:
- Named enterprise customers in the headline or summary.
- Prior consulting stints (Palantir FDSE, Deloitte, Accenture, McKinsey Digital).
- Regulated-vertical exposure: banks, insurers, gov, healthcare, defense.
- Public writeups or talks about a specific deployment.
3. Drop the remote filter
Remote FDE roles are structurally rare because the job requires embedding on-site with customers up to 50% of the time. If your req is remote-only, you are fishing in the 5% of the market that will churn out in 12 months. Rewrite the JD to say "hybrid, 50% travel" and your response rate will climb.
4. Stop screening for Python skill tags
Only about 12% of FDE-titled US professionals list Python on their profile in Refolk's index. Screen for it in the phone screen instead. If you gate on the skill chip, you cut 88% of qualified people.
5. Mine the long tail
Google and Deloitte account for ~40% of postings. The other 130+ companies each hire 1 to 3. That long tail is where wage arbitrage lives, because those hiring managers do not have a comp band cemented yet. It is also where FDE recruiting 2026 gets won or lost, because the sourcers who understand the JD nuance close faster than the ones running a stock SWE playbook.
For teams doing forward deployed engineer sourcing at any real volume, Refolk collapses steps 1 through 4 into a single plain-English query. You describe the profile, the tool searches across GitHub, LinkedIn, and the open web, and it returns ranked people whether they call themselves FDE, Deployment Strategist, or Applied AI Engineer.
The named companies to watch
The FDE market in 2026 is defined by roughly a dozen employers on the demand side and a long tail on both sides. The ones worth tracking:
- Palantir Technologies. Coined the role in 2011; still the largest single employer of FDE-titled people in Refolk's index. Best poaching target on comp.
- OpenAI. Actively hiring FDEs in NYC with global footprint across Seattle, DC, London, Munich, Paris, Singapore, Tokyo, Sydney, Dublin. Highest TC in market.
- Anthropic. $200K to $300K base postings, widely reported to be firm on offers.
- Google Cloud. Thomas Kurian confirmed the FDE ramp; 34 active postings makes Google the single largest active buyer.
- Deloitte. 19 active postings; the consulting-major entry point into FDE hiring, and useful evidence the role has jumped the AI-native fence.
- Long-tail startups in Refolk's index. Cresta, Modal, Roboflow, Gecko Robotics, Amp Code, Scale AI. Quietly hiring, less crowded pipelines.
Aaron Levie put the demand side plainly: FDEs are becoming a fundamental component for successful corporate AI rollouts. That is why Google's 34 reqs is not an anomaly. It is a leading indicator.
FAQ
What is the difference between a Forward Deployed Engineer and a Solutions Engineer?
Functionally, in 2026, often nothing. Both embed with customers to ship product against real deployments. The naming diverges by company culture: Palantir and OpenAI use "Forward Deployed Engineer" to signal engineering seniority, while enterprise-SaaS-native companies use "Solutions Engineer" or "Deployment Strategist" for the same headcount. In Refolk's index, the sibling-title cluster is 3.03x the exact FDE pool, and the top employer is the same (Palantir) across both.
Is Forward Deployed Engineer a sales role?
No. In an analysis of 1,000 FDE postings, 0% mentioned a sales quota. FDE is engineering with deep customer exposure, not pre-sales SE and not account management. The confusion comes from the on-site component and the customer-facing surface area, but the deliverable is shipped code inside the customer's environment, not signed contracts.
Why is NYC beating SF for FDE hiring?
Regulated-industry demand. NYC holds 35% of FDE postings versus SF's 11% because fintech, insurance, and compliance-heavy buyers need engineers who can deploy inside SOC 2 and PCI boundaries, and those buyers cluster in New York. It is a fintech flip, not a tech-industry flip, and it means the ideal candidate profile skews toward financial-services or gov exposure rather than pure AI-lab pedigree.
How do I compete with OpenAI and Anthropic on FDE comp?
You do not, on cash. OpenAI's median FDE TC is around $555K and Anthropic packages run $350K to $550K. Compete on scope and customer surface: a Series B founder can offer an FDE the first ten customers, direct product influence, and a comp package in the $250K to $350K TC range that a labs FDE will actually consider if the equity is real. Palantir FDSEs at $215K median TC are the most winnable target in the market.