How to become a forward deployed engineer is mostly a question of evidence. To become forward deployed, you need to show that you have built something for a real customer, scoped it honestly and handed it over. Employers read that record before they read a badge. A credential helps once the portfolio exists.
Key takeaways
- What the forward deployed engineer role involves and where it came from.
- Which skills job adverts ask for, and how to prove them with deployment case studies.
- Where a credential fits once the portfolio is in place.
What it takes to become forward deployed
According to Wikipedia, a forward deployed engineer is a customer-facing software engineer who helps implement their employer’s product inside a client company, often working alongside the client’s employees for a defined period. The Pragmatic Engineer reports that Palantir created the role in the early 2010s and called it “Delta”.
The word that matters is engineer. The First Round Review draws the line this way: unlike a solutions consultant or sales engineer, the FDE still writes and debugs production code. The job is the last mile of a product, built inside the customer’s own systems and politics.
Expect time on site. The Pragmatic Engineer says Palantir expects around 25% of an FDE’s time with customers, while healthcare AI company Commure estimates up to 50%. Databricks lists travel to customers at 20% of the time in one posting. OpenAI’s posting describes the scope as discovery, technical scoping, system design, build and production rollout. That is a whole deployment, not a single ticket.
How big the market is and how to read the numbers
Demand is real, but the headline percentages need care. Lightcast reports US postings rising from roughly 200 in 2024 to around 1,200 in 2025, with more than 5,200 in the first seven months of 2026. The same Lightcast piece notes that demand remains small compared with established technology occupations.
KDnuggets found growth estimates ranging from 350% to 1,165% depending on who counts and when, and points out that the Indeed figure is an index against a small base on one job board. Treat the growth as a direction, not a guarantee of openings.
Pay is part of the appeal. According to Fortune, citing Lightcast, median advertised salaries for forward-deployed engineers were above $188,000, against roughly $145,000 for traditional software engineers. Those are US figures, so read them as a signal rather than a UK expectation. Wikipedia also records that some engineers find the role undesirable because of travel and the pressure to solve customer problems quickly.
The skills job adverts actually ask for
Entry bars vary widely. OpenAI’s posting asks for five or more years of engineering or technical deployment experience that includes customer-facing work, plus production-grade Python or JavaScript and experience building systems on LLMs. Google Cloud’s Applied AI role asks for five years of Python or similar and experience architecting AI systems on cloud platforms. Scale AI’s GenAI posting says at least two years of relevant experience is preferred.
There are also routes in with less. The Pragmatic Engineer notes that Palantir hires people with as little as a year of post-college work experience, while Commure and Matta value real-world experience of building and shipping projects from start to finish. Salesforce says it tripled its FDE team in six months by drawing on engineering, professional services and customer success, and that around 40% to 50% of FDE movement there is internal.
Three skill groups to evidence
Across those postings, the skills fall into three groups, and a portfolio should cover each.
- Build: production code in Python or a comparable language, cloud basics, data pipelines, and LLM application work including retrieval and evaluation.
- Discover: finding the real problem behind the stated one, and saying no to work that does not serve it.
- Hand over: documentation, training and a system the customer can run without you.
The First Round Review reports that Palantir’s FDEs passed the same interview loops as software engineers, and that its FDE interviews were built around real customer technical challenges. Prepare for both halves.
Build a portfolio that shows deployments, not demos
A notebook that calls an API is a demo. A deployment has a user who was not you, a workflow that existed before your system, and a point where you stopped being needed. Start small: a local business, a charity, a team inside your current employer. Pick one repeated task, such as triaging inbound emails or reconciling spreadsheets, and ship something that people use for a few weeks.
Write each project as a case study that follows the order a deployment actually runs.
AIU teaches the deployment lifecycle in seven stages: post-sale handoff, discovery, scoping, build, deploy, iterate and hand-back.
Two stages deserve most of your words. In discovery, record who you interviewed, what they told you that the brief did not, and how it changed the plan. In hand-back, record what the customer’s team now does without you, and what you left behind to make that possible. Employers see plenty of build sections. Few candidates can show either of these.
Two questions help you test each case study before you publish it. Apply the Refusal Test: your write-up should name something the system declined to do and show what that refusal looks like to the user.
What did your system refuse to build, and what does the refusal look like?
Then apply the Signature Test, which checks that someone with authority accepted the risk of going live.
Who signs the risk statement before go-live?
Finish each case study with the measures the role is judged on.
AIU teaches that Forward Deployed Engineers are measured on three things: deployment velocity, customer success and product feedback quality.
That framing matches how employers describe the job. OpenAI’s posting says FDEs measure success through production adoption, measurable workflow impact and eval-driven feedback that changes product and model roadmaps. Use your own numbers where you have them, and state plainly where you do not.
Where a credential fits, and where it does not
The order matters. A Salesforce leader quoted on the company’s news site says it is “curiosity over credentials” there. That is one employer, but the pattern is common in postings: they ask for experience and shipped work, not a certificate.
A credential still has a place. It gives a hiring manager a second, independent signal alongside your case studies, and it gives you a structured way to find gaps. For readers who want that, the Certified Forward Deployed Engineer (CFDE)™ is one route, best sat once there are deployments to bring to it. Afterwards, CFDE sits on a CV beside the portfolio, not in place of it.
The portfolio also protects against a market problem. Wikipedia notes that Andreessen Horowitz called the title “title arbitrage”, a way to signal importance for roles previously called solutions engineers or integration engineers. Titles are inconsistent, so a candidate who can show real deployments is easier to place than one who can only show a label.
A practical route from where you are now
- Secure the engineering base. Be able to write, test and deploy code in one backend language and one cloud, and to work with messy data.
- Ship one small deployment. Use a real user and a real workflow, and keep it running long enough to learn something.
- Write it up in lifecycle order. Include the discovery notes, the scope you cut, the failure you hit and the hand-back.
- Search under several titles. Wikipedia lists overlap with solutions architects, sales engineers, customer engineers, professional services engineers and systems integrators, so adverts use varied wording.
- Prepare for case-style interviews. Practise taking a messy customer problem and saying what you would ship first and what you would refuse.
If you already work in professional services, support engineering or data engineering, the shortest path may be an internal one, as Salesforce’s figures suggest. Build the case study inside your current job first.
When the portfolio has two or three solid case studies, a recognised credential can sit alongside them.
Explore the CFDE Foundation programme →
Frequently asked questions
Do I need Palantir experience to become a forward deployed engineer?+
No. Palantir created the role, but other employers recruit from elsewhere. Salesforce says it built its FDE team from engineering, professional services and customer success staff, and it reports that a large share of FDE movement there is internal. A portfolio of real deployments is the better preparation.
How is a forward deployed engineer different from a solutions engineer?+
A forward deployed engineer still writes and debugs production code inside the customer’s environment. According to the First Round Review, that separates the role from a traditional solutions consultant or sales engineer. The titles overlap in practice, so read the responsibilities in each advert rather than the job title.
Do I need a certification to get a forward deployed engineer job?+
The postings read for this article, from OpenAI, Google Cloud and Scale AI, asked for engineering experience and shipped work rather than a certificate. A credential can add an independent signal, but it works best after you have deployment case studies to show.
