The forward deployed engineer vs solutions engineer question comes down to one thing: whose production does the code run in, and who answers when it breaks? A forward deployed engineer builds inside the customer’s environment and stays accountable after go-live. A solutions engineer proves fit before the sale. An AI engineer usually builds the vendor’s own product.
Key takeaways
- The one question that separates a forward deployed engineer from a solutions engineer, solutions architect and AI engineer.
- Where the role came from and why AI labs and enterprises are hiring for it now.
- How the work is measured, handed over and checked for real ownership.
Forward deployed engineer vs solutions engineer: the line is production
A forward deployed engineer (FDE) works inside the customer’s organisation to make a system run in that customer’s live environment. According to Gainam, a solutions engineer operates pre-sale to win the deal, while an FDE operates post-sale to ship the deal into production. The same page sums up the split in one line:
The SE proves it can work; the FDE makes it actually work.
The practical differences follow from that. Gainam’s comparison puts the solutions engineer in a demo or sandbox with sample data, and the FDE on the customer’s live infrastructure with real data. It also points to the 2 a.m. test: when something breaks, the SE has moved on and the FDE is on call.
Titles are unreliable, so read the duties. Phaedra Solutions offers three checks: whether the work has to become production code, whether the bottleneck sits before or after technical validation, and what outcome the person should own. Markaicode notes that labels blur, with some solutions engineers shipping production code and some FDEs supporting pre-sales. Phaedra adds that strategic or technically complex deals can need an FDE before the contract is signed. The test is what the person is held responsible for, not when they join.
Where the forward deployed engineer role came from, and why demand has grown
The Pragmatic Engineer reports that the role was created at Palantir in the early 2010s and named Delta, and that until around 2016 Palantir had more of them than conventional software engineers. Dating varies by source. Atlassian says Palantir invented it in the mid-2000s, when early intelligence and defence customers could not fully explain their problems or share their data, and the answer was to send engineers to live inside those organisations, build against real conditions and feed what they learned back into the product.
That last part matters. In the First Round Review, Palantir alumna Shilpa Balaji argues that pure implementation or configuration has drifted from the original role. An FDE discovers things in the customer’s context and sends them back to the product team.
Demand has followed AI adoption. A Nasscom community post reports FDE postings grew roughly 729% year on year by April 2026, with AWS, Microsoft, OpenAI, Anthropic, Google and Palantir building units around the role. Per Forbes, Anthropic announced a services joint venture with Blackstone, Hellman & Friedman and Goldman Sachs on 4 May, and a week later OpenAI announced a Deployment Company capitalised at more than $4 billion, acquiring Tomoro for its roughly 150 deployment engineers. Forbes also cautions that the economics of the FDE model are unproven outside Palantir.
The labs also hire directly: Anthropic’s careers page lists Forward Deployed Engineer openings in Munich, Paris, London and New York.
How the solutions engineer, solutions architect and AI engineer differ
Solutions engineer
The solutions engineer is a mature role with a clear job: technical validation that helps a deal close. According to Hayat, if a person’s pay moves with a deal closing rather than with something going live, the hire is pre-sales whatever the title says. Account count is another giveaway. Valletta Software argues that forward deployment goes deep on one to three customers, and that an FDE covering fifteen accounts is really a solutions engineer.
Solutions architect
The architect owns the design. Hayat describes the architect as rewarded for design and advice, and the FDE as rewarded for execution and holding the pager when the design meets the customer’s real data. An excellent architecture and a working deployment are different things, and the gap between them is the FDE’s job.
AI engineer
“AI engineer” covers a wide range, so apply the same question. An AI engineer working in the vendor’s own codebase sits on the software engineer side of the line. Rocketlane puts it simply: software engineers build the product, and FDEs make it work for a particular client in a particular environment. At the labs the titles also overlap. According to Szymon Paluch’s guide, an FDE builds in the customer’s systems, while Anthropic’s Applied AI Architect is a pre-sales role. Read the duties, not the label.
How a forward deployed engineer is measured and handed over
Gainam says solutions engineers are measured on pipeline and win rate, while FDEs are measured on whether the deployed system ships and delivers ROI. The scoreboard shapes behaviour, which is why the roles feel so different day to day.
AIU teaches a set of measures for the FDE role in particular.
AIU teaches that Forward Deployed Engineers are measured on three things: deployment velocity, customer success and product feedback quality.
Ownership also runs across a sequence of work, and handovers are where it tends to fail.
AIU teaches the deployment lifecycle in seven stages: post-sale handoff, discovery, scoping, build, deploy, iterate and hand-back.
The final stage is the one a solutions engineer never reaches. Hand-back means the client’s own team runs, monitors and changes the system without the builder in the room. If the system only works while its author is present, it has not been handed back, whatever the go-live report says.
Two checks help separate owners from advisers on any deployment.
Who signs the risk statement before go-live?
The question forces a name onto the risk. A person who writes the risk statement owns the outcome, and a person who only presented the plan does not.
Five tasks the client team performs while the deployment engineers observe.
It tests the client team, not the vendor. If the client cannot perform the tasks, the build is not finished.
Which role to hire, or train for
For a hiring manager, Hayat’s rule is to count unshipped deployments: if there are none and the pipeline is thin, hire a solutions engineer; if signed customers are not yet live, hire a forward deployed engineer.
For an individual, the move from solutions engineer to FDE is common. What has to change is the standard of the code, because a demo environment forgives things production does not, and the relationship with failure, because an FDE who ships a broken integration still owns it on Monday.
The Certified Forward Deployed Engineer (CFDE)™ was built as the first certification standard for the Forward Deployed Engineer role. The course teaches, and a separate examination administered by AIU certifies.
If your code will run in a customer’s production, the next step is to see how AIU teaches the role.
Explore the CFDE Foundation programme →
Frequently asked questions
What is the difference between a forward deployed engineer and a solutions engineer?+
A solutions engineer works before the sale, using demos and proofs of concept to show the product fits. A forward deployed engineer works after the sale, writing production code in the customer’s environment and staying accountable for it. Gainam’s comparison lists owning the production outcome as a clear no for solutions engineers and a yes for FDEs.
Is a forward deployed engineer the same as an AI engineer?+
Not by default. An AI engineer working in the vendor’s product codebase ships to all customers, while a forward deployed engineer builds inside one customer’s systems. At AI labs the titles overlap, so check whether the duties include production work in the customer’s environment. Szymon Paluch’s guide notes that the two titles overlap at the labs.
Can a solutions engineer become a forward deployed engineer?+
Yes, and Hayat calls it one of the most common routes into the role. The main changes are a higher standard of production code and accepting ownership of failures after go-live, rather than moving on to the next prospect.
