A forward-deployed engineer is a software engineer who works inside one customer's team and systems, scopes the problem, builds the software and keeps it running in production. Most mid-size businesses don't need to hire one. They need what one delivers, and renting an embedded engineering team for a defined engagement is usually the more realistic way to get it.
What is a forward-deployed engineer?
A forward-deployed engineer (FDE) is a software engineer assigned to one customer who learns how the work actually runs, builds software inside that customer's systems and stays until it works in production. The job ends when the software works on real work.
Palantir made the role famous. It calls these engineers Deltas and describes the difference from its product engineers this way: a product engineer's focus is "one capability, many customers," while a Delta's is "one customer, many capabilities." Palantir also says a Delta "does significantly more engineering and technical work than a consultant."1
Companies that use AI are adopting the model too. Forrester reports that earlier in 2026 Uber sent 30 of its most AI-proficient engineers to sit inside finance, legal and HR, and that financial pacing reports dropped from two days to 10 minutes.2
Why is everyone talking about forward-deployed engineers now?
Because the biggest AI providers are betting on them. Forrester counts roughly $9B committed by four providers, Anthropic, OpenAI, Amazon Web Services and Microsoft, between May 4 and Jul 15, 2026, to put engineers alongside customers.2
Most of that is aimed at the largest companies. Anthropic's new enterprise AI services firm with Blackstone, Hellman & Friedman and Goldman Sachs is the exception: it says it will bring AI to mid-size companies, starting with the investors' portfolio companies.3 In our view, if you're a smaller customer of a big provider, it's hard to get real customization.
How is a forward-deployed engineer different from a consultant, a solutions engineer or a software vendor?
The difference is who builds the system and who is still there when it breaks. A consultant hands you a plan, a solutions engineer shows you a proof of concept, and a forward-deployed engineer builds the working system and supports it in production.
| When they work | What you get | Who runs it after launch | |
|---|---|---|---|
| Strategy consultant | Before anything is built | An assessment and a recommendation | Someone else |
| Solutions engineer | Before the sale | A demo or proof of concept | Not their job |
| Vendor support | After the sale | Training and help with the product | Your team, with tickets to the vendor |
| Forward-deployed engineer | After the sale, inside your team | A working system in your own tools, plus documentation | The engineer, alongside your team |
One caution. When the engineer comes from a platform vendor, the work tends to pull you deeper into that platform. Forrester puts it plainly: "deep integration creates its own gravity," and as data and workflows build up around a platform, the cost of moving rises.2 Ask where the code and data will live, and who owns them.
Does a mid-size business need a forward-deployed engineer?
A 50 to 2,000 person business usually needs what a forward-deployed engineer delivers more than it needs one on the payroll. Renting an embedded team for an engagement gets you working systems without competing with the largest AI companies for scarce engineers.
Hiring is expensive and slow. Self-reported pay on Levels.fyi puts median total compensation for a forward-deployed engineer at $220K (checked Nov 3, 2026),4 and one engineer can't cover every system, workflow and on-call shift alone.
| Hire your own | Rent an embedded team | |
|---|---|---|
| Best when | You expect a steady stream of custom AI projects for years | You have a few high-value workflows to build and run |
| Time to start | A full hiring cycle | Weeks |
| Coverage | One person's skills and hours | A team with several specialties |
| What to check | Can you keep the role interesting enough to retain them? | Do you own the code and data, and does the team stay after launch? |
If you rent, vet the firm the way you would any partner. Our guide on how to choose an AI consulting firm lists the questions to ask before you sign.
What should you expect from an embedded engineering team week to week?
Expect a few weeks of listening, then fast building, then ongoing improvement. The real process is rarely the one in the documentation, so a good team starts by watching the work.
- First two weeks: learn how the work runs. The engineers meet the team and interview the people doing the work. By the end of week two there's a clear scope and goal.
- Next few weeks: build the first workflow. Pick one with return you can attribute directly. Your team reviews the output and keeps the approvals.
- After launch: monitor and improve. The engineers watch how it performs on real work, fix what breaks and move to the next workflow.
Five questions to ask any firm that offers forward-deployed engineers:
- Who writes the code, and who is on call when it breaks?
- How many accounts does each engineer cover at once?
- What does my team own at the end? The answer should be everything: code, data and documentation.
- Whose platform is it built on, and what would leaving cost?
- When the build is done, do you leave or stay? On discovery calls, prospects often tell us they were burned by a vendor who built something and left.
How do Soren's engineers embed with a client team?
At Soren AI, you get the whole engineering team, typically 10+ engineers, not one person assigned to your account. The first two weeks are on site or virtual, whatever makes sense, and we're building in the background while we learn. You own everything we build.
Engagements usually run a year or more, and our engineers stay after launch to monitor, support and improve what they built. At one law firm, client intake went from three days to 20 minutes. More on how we work, or see what happens in an AI transformation assessment.
Talk to the engineers
Start with a 30-minute call with our team. We'll look at where your back office loses time and whether an embedded team is the right fit.
Frequently asked questions
Is a forward-deployed engineer the same as a solutions engineer?
No. A solutions engineer works before the sale, building demos and proofs of concept to show what a product can do. A forward-deployed engineer works after the sale, inside your team, building the production system and supporting it once it's live. Some companies blur the titles, so ask what the person actually delivers.
Do forward-deployed engineers work on site?
Often, but not always. The point is to work where the work happens, so many engineers spend time on site early on to interview staff and watch the process. After that, much of the building can happen remotely, with regular visits. Ask any firm how it splits on-site and remote time for your project.
What happens when a forward-deployed engagement ends?
It depends on the firm. Some hand over code and runbooks and leave, so your team must run the system from day one. Others stay to monitor and improve what they built, then transition gradually. Before you sign, get the exit terms in writing: what you own, who supports it and how long.
Can a 200-person company work with forward-deployed engineers?
Yes, if the firm is set up for mid-size clients. Ask how many engineers you'd get, how many other accounts they cover and whether they stay after launch. The right firm gives a 200-person company the same model large enterprises use: engineers who learn the workflow, build inside existing systems and keep improving them.
Sources
- Dev versus Delta: Demystifying engineering roles at Palantir, Palantir Blog
- Forward-Deployed Engineers Are The Training Wheels For AI Reinvention, Forrester, July 30, 2026
- Anthropic Partners with Blackstone, Hellman & Friedman, and Goldman Sachs to Launch Enterprise AI Services Firm, Business Wire, May 2026
- Forward Deployed Engineer Salary, Levels.fyi, checked November 3, 2026
