Forward Deployed Engineer
Move into a business. Understand it from the inside and out. Then build what it actually needs — with a fleet of AI employees working under you. Remote. Senior. You own the outcome.
The role, plainly
The job was invented at Palantir. Rather than sell software and leave, they sent engineers to live inside the customer's business — sitting with the people doing the work, learning the problems nobody had written down, and building against them on the spot. It worked so well that the industry copied it.
What changed recently is the leverage. An engineer's output used to be capped by how fast they could build. That cap is gone: at Pruta every engineer directs a fleet of AI employees that carries the production work around the clock. So the scarce skill is no longer typing. It is knowing which problem is worth solving, and being close enough to a business to see it before anyone reports it.
That is what we hire for, and it is why this job looks less like a normal engineering role and more like being handed a business and asked to make it work better.
What you would actually do
- You move into a client's business. Their systems, their calls, their inbox, their numbers. You learn how the work actually flows rather than how the org chart says it does — and you keep learning it for as long as the engagement runs.
- You find the problems nobody reported. The leak in the sign-up flow. The customers who quietly stopped receiving email. The process eating a day a week that everyone stopped noticing years ago. Nobody files a ticket for these. Finding them is the job.
- You build the fix — with a fleet under you. You direct a team of AI employees that carries the production work: writing, wiring, testing, shipping. Your leverage is not how fast you type. It is knowing exactly what should exist and being able to get it built by Friday.
- You answer for the result. Not for a deliverable, a ticket, or a sprint. You are the person the client calls, and the one accountable for whether their business is measurably better than it was last month.
Whether this is your job
You are probably a fit if…
- You have shipped real things to real users and can point at them.
- You already work with AI models daily and get far more out of them than most people do.
- You can sit with a non-technical owner and leave understanding their business better than they can explain it.
- You are comfortable deciding what to build, not just how to build it.
- You would rather own an outcome than be handed a specification.
- You can write plainly — to a client, in a report, in an email that lands.
You are probably not, if…
- You want a well-defined ticket queue and a product manager between you and the customer.
- You prefer to go deep on one codebase for years.
- Client contact sounds like a distraction from the real work.
- You are looking for a title and a ladder more than a problem and a free hand.
- You would rather not be the one accountable when something does not work.
What we offer
- $120,000 to $1,000,000 a year. The floor is what a senior engineer earns. The ceiling is what someone who can carry several client engagements, with a fleet working under them, is genuinely worth to us. Where you land depends on what you can actually run — and we talk about it openly and early, not at the end.
- A fleet under you from day one. You are not a lone contractor. AI employees handle production work under your direction, with the infrastructure, monitoring, and cost controls already built.
- Remote, and genuinely so. The whole company works this way. We care where your head is on the client's problem, not which chair you are sitting in.
- Real autonomy, real accountability. You decide what gets built. You also own whether it worked. If that trade sounds good rather than frightening, this is the right room.
- Range most engineers never get. You will see inside businesses most people only read about, and you will touch every part of the problem — from what to build, to shipping it, to whether the client's numbers moved.
How we hire
No CV screen. Send links to things you have built and tell us what changed for the business because you built them. We read that first, and often that is enough to know.
One real conversation. Not a puzzle round. We talk through a business we have actually deployed into, and you tell us what you would have looked at first.
A paid trial deployment. Where it makes sense, we put you on a small real piece of work and pay you for it. Both sides learn far more from a week of the actual job than from another interview.
Apply
Tell us what you have built and what it changed. A human reads every application.