Three years ago almost nobody put "Forward Deployed Engineer" in a job posting. Today Google, Anthropic, ByteDance and Accenture's Palantir practice do, along with dozens of AI startups deploying inside client environments. The role has moved from Silicon Valley curiosity to an actively hired position in London, and many senior developers are looking at it sideways without being sure what it involves.
Straight to the point. This is the map of the role: what an FDE does day to day, which skills the market actually asks for, where the people landing in it come from, and why UK demand has picked up pace, backed by measured market data rather than trend headlines.
TL;DR. A Forward Deployed Engineer (FDE) is an engineer deployed alongside the client to build and integrate software, today almost always AI-based, inside their real operation. It mixes development, integration and direct business contact. An FDE writes production code and turns a business problem into a system that works. According to Shakers' market analysis, a search for "forward deployed engineer" across UK tech returned 83 active postings in September 2026, 65 of them in London.
What does a Forward Deployed Engineer do?
A Forward Deployed Engineer builds and integrates software directly in the client's environment, side by side with the people who will use what they build and inherit its bugs if something breaks. The name says it: deployed forward, all the way to where the business happens. In practice it combines three things that used to be separate. They write production code. They integrate that code with the systems the client already runs. And they talk to the business to understand what people actually need, beyond what a ticket says. That last part, translating what the business asks for into what the system truly has to do, is where an FDE earns their salary and where a purely technical profile tends to fall halfway.
That mix is the differentiator. A classic developer receives a specification and implements it. An FDE arrives earlier: sits with the client, uncovers the real problem, prototypes a solution and takes it to production, iterating on the ground. In the AI era the pattern has exploded, because deploying an agent or a language-model system into a company's operation demands exactly that hybrid profile: someone who understands the model and the business process it will touch. Both at once.
Why is the role growing in the UK?
Because AI left the lab and entered operations. When a company wants an agent resolving support cases or a system automating a real workflow, it needs someone to build it inside their house, with their data and their systems. That someone is the FDE.
The data backs it. Shakers' market analysis counted 83 active UK postings titled Forward Deployed Engineer in September 2026, spread across 56 distinct employers. Alongside the AI labs, consultancies and integrators hire heavily: Accenture runs a Palantir FDE practice, Salesforce, Kyndryl, ServiceNow and Databricks all post FDE roles in London, and startups from ByteDance's BytePlus to seed-stage teams hire the same profile. Advertised senior packages range from £90K to £250K, with two postings at £160K-£310K. One signal says more than any trend report: the UK government's No10 Innovation Fellowship runs its own Forward Deployed Engineers programme, embedding FDEs across the public sector. When a role reaches Downing Street, it has stopped being a fad.
Which skills and stack does an FDE need?
The base is engineering. An FDE has to be solid building and deploying software to production; publishing research on model architectures is a different job, and knowing the difference marks from day one which profiles fit the role. Looking at the real UK postings, the pattern repeats: Python leads, with TypeScript and Go close behind for the product side. On top sit the clouds (AWS, Azure, GCP), containers (Docker, Kubernetes) and frameworks for working with language models such as LangChain or LlamaIndex, plus the OpenAI and Anthropic APIs. Nothing exotic.
Here is the nuance. The skill that truly separates an FDE from a good developer sits outside that technology list: the ability to sit with a client who cannot quite explain what they want, understand the real problem underneath and decide what to build with their own judgement. The technical part can be learned. Fewer people have the other one.
How do people become Forward Deployed Engineers?
There is no single path, and that is good news. The profiles that end up as FDEs converge from three quite different starting points. Full-stack developers tired of receiving closed specifications who want to sit closer to the business. Pre-sales and solutions engineers who already talk to clients and decide to go down to the code. And data engineers or scientists who realise their model only matters if someone actually deploys it. Three doors, one destination.
What all paths share is the attitude: wanting to solve the client's whole problem rather than your slice of the ticket, and being comfortable sitting in front of the business asking whatever it takes to understand what to build. That part cannot be faked.
FDE on staff or by project?
The role was born inside product companies, but it fits at least as well in project format. A company deploying AI rarely needs a permanent FDE: it needs that capability during implementation, and then the knowledge to stay with the internal team. That is where certified talent collaborating by project makes sense. The professional comes in, builds alongside the internal team and demonstrates their skills on the project itself, rather than selling them on a CV.
If that route fits you, the projects page for Forward Deployed Engineers at Shakers explains how it works and how to chain engagements without depending on a single employer.
Frequently asked questions about the Forward Deployed Engineer
- What is the difference between an FDE and a solutions engineer?
-
A solutions engineer usually supports the sale and the design, and rarely writes the code that reaches production. An FDE does. They stay after the sale, build the solution inside the client and leave it running. There is overlap in talking to the business; the FDE closes the loop all the way to production code.
- How much does a Forward Deployed Engineer earn in the UK?
-
Advertised senior packages in the September 2026 UK postings range from £90K to £250K, with senior roles at AI labs and Palantir practices at the top of the band. As with every senior role, the exact number depends on seniority and company type.
- Do you need to be a machine learning expert to be an FDE?
-
You do not. An FDE needs solid engineering (Python, cloud, integration) and fluency using language models through their APIs and frameworks, with training models from scratch as a different job. The value sits in deploying AI that works inside the client's operation; other profiles own the research side.
- Is it a good time to move into this role?
-
Market signals say yes. Shakers' market analysis counted 83 active UK Forward Deployed Engineer postings in September 2026, published by AI labs, consultancies and startups alike. It is a young role with more demand than prepared profiles, which favours whoever positions early.
The Forward Deployed Engineer is a structural consequence of AI leaving the experimentation phase: companies want it working inside their operation, built by someone on that ground rather than in a demo. If you are a senior developer, or you come from pre-sales or data, the practical conclusion is simple: the market is already searching for this profile before you go searching for it, and that window of advantage for early movers rarely stays open long.
At Shakers we work with certified talent in AI skills collaborating on real client projects, exactly the FDE's territory. If you want to take on projects of this kind with your skills already proven in real work, the Forward Deployed Engineer projects page lays out what is available and how certification works.
