Embed.
Sit with the customer's team, often on site, until you can argue about the workflow.
The forward deployed engineer handbook.
What forward deployed engineers do, what they earn, and how they run parallel engagements.
What is a forward deployed engineer?
A forward deployed engineer (FDE) is a customer-facing software engineer who embeds with a client organization and ships production software inside the customer's own environment, owning the outcome end to end, post-sale, until it works. Palantir coined the role around 2010; the AI deployment wave made it the fastest-growing job in software.
The name borrows the military's “forward deployment”. By roughly 2016 Palantir employed more forward deployed engineers than traditional software engineers, per The Pragmatic Engineer. Enterprise AI's bottleneck has since moved from model capability to deployment, and a16z called it “the hottest job in startups” in 2025.
What does an FDE actually do? Five phases. Only one is writing code.
Discover.
Map the systems of record, the wrong data, the undocumented process. FDE interviews test this.
Build.
Production code in the customer's stack and repos. Not a demo.
Deploy.
Through their review, their compliance, their release process.
Feed back.
Turn what the field taught you into product. This is what separates an FDE from a consultant.
FDE vs solutions engineer vs consultant. Who owns the outcome.
Solutions engineers sell it.
Pre-sale demos and POCs, handed off at contract, with deal notes as the feedback loop.
Consultants spec it.
Post-sale work to spec, per SOW, ending in a final report when the SOW ends.
Forward deployed engineers ship it.
Production code in the customer's stack, owned end to end until it works. Learnings become product.
Why now. Four numbers that explain the moment.
+1,165% FDE job postings, Jan–Oct 2025 vs 2024 (Bloomberry)
$4B behind OpenAI's Deployment Company, May 2026
$1B AWS Forward Deployed Engineering program, June 2026
~$174K median posting salary, approximate (Bloomberry)
What 1,000 job postings ask for. An engineer who ships AI systems, not a salesperson who codes.
Python 66%, TypeScript 35%, AI agents 35%, AWS 32%, LLM experience 31%, RAG 12%. From Bloomberry's analysis of 1,000+ FDE postings.
Forward deployed engineer salary, 2026.
- Glassdoor~$156K average · $125K–$198K (25th–75th percentile)
- Bloomberry (1,000+ postings)$173,816 median · 70% mention equity · 0% carry quota
- Frontier AI labs$350K–$450K total compensation
Why is the FDE workflow so hard to scale?
Three to five parallel engagements is the normal load, each with its own repository, environment and stakeholders. Nothing about that composes.
Coding agents made generating code cheap. What remains is running it, reading the diff and getting it through the customer's review. Agents produce reviewable output faster than one person can consume it, so an FDE's throughput is now bounded by review attention, and a one-terminal work surface cannot carry a parallel workload.
The FDE tool stack in 2026. Three layers are well served. One is missing.
- BuildA coding agent (Claude Code or Codex) plus an eval harness.
- DeployThe customer's infrastructure; Modal, Vercel or Fly when the FDE owns the target.
- IterateTelemetry from the deployed system, feeding the next change.
- Work surface (the gap)Where every engagement is tracked, reviewed and shipped in parallel. Nobody built it.
Call it the engagement board: every client a lane, every unit of work a ticket with its own branch and agent. Running five engagements in parallel walks the loop, and the 2026 FDE tool stack audits which layers survive a handoff.
How FDEs run parallel engagements with Fredrin. One board, every client.
One project per client.
Each one pointed at that customer's repository.
A branch and an agent per ticket.
Four engagements run at once instead of queuing behind one terminal.
Review, then ship.
Work lands in Review: read the diff, ship it or send it back.
The answer to the security review.
A desktop app running Claude Code or Codex on the subscription you already pay for. A client's code and your agent conversations stay on your machine unless you turn that on.
See how the board works, or compare it with Cursor, Devin and Linear.
Frequently asked.
Which companies hire forward deployed engineers?
Palantir invented the role around 2010. Today OpenAI, Anthropic, AWS, Google, Ramp, and hundreds of AI startups hire FDEs. Most postings come from companies of 11–200 people.
Is a forward deployed engineer just a consultant?
No. Consultancies deliver to spec and leave when the engagement ends. FDEs apply the product model to embedded work: what the field teaches becomes platform features, not a report.
Is FDE a good career move in 2026?
Postings grew roughly 1,165% year over year, OpenAI and AWS funded FDE programs with billions of dollars, and median posting pay clears $170K (approximate). It is the clearest fast lane between software engineering and the AI deployment wave.
How much do FDEs travel?
Typically 20–50% of the time. The “forward deployed” part is literal. Some AI-lab roles are remote-first with periodic customer on-sites.
What is the difference between a forward deployed engineer and a sales engineer?
A sales engineer works the pre-sale: demos, proofs of concept, and technical objection handling, usually against a quota. An FDE picks up after the contract is signed and writes production code in the customer's stack. Roughly 0% of FDE postings carry a sales quota, which is the cleanest way to tell the two roles apart in a job description.
What tools do forward deployed engineers use?
A coding agent (Claude Code or Codex) plus an eval harness for the build loop; the customer's own cloud (or Modal, Vercel, or Fly when the FDE owns the target) for deploy; and telemetry from the deployed system to drive the next iteration. The missing layer is the work surface that tracks all of it across several engagements at once.
How many customer engagements does one FDE run at once?
Three to five is typical, and that is the real constraint on the role, not how fast anyone writes code. Each engagement is a separate repo, environment, and stakeholder, so the scarce resource is the FDE's review attention, which coding agents consume faster than they free.