Article

What is a Forward Deployed Engineer? Role, skills and daily work

What is a forward deployed engineer? Explore forward deployed engineering, daily work, skills, jobs and salary questions, with real examples from my work.

Published 17 November 20267 min readFDEProductCareers

A Forward Deployed Engineer is an engineer who enters a customer's context, builds software in their real systems and brings what they learn back to the product team. It is not simply a person who presents a demo or collects requests. The value comes from connecting a concrete need, engineering work and product decisions. Here I explain the role through my approach to software and documented experience, without claiming that my previous jobs already carried this title.

Forward deployed engineering: connecting the problem to the product

Deployed does not necessarily mean sitting in the customer's office every day. What matters is closeness to the problem: understanding who uses the system, which data passes through it and where a technical choice changes the work. A seemingly small feature may require talking to people responsible for orders, content or integrations. The engineer does not treat those conversations as an obstacle before coding. They help decide which code is worth writing and whether the result will actually be usable.

The product connection distinguishes this approach from customisation for its own sake. A request may reveal a general limitation, an incomplete configuration or a need specific to one customer. Feeding it back does not mean promising it a place on the roadmap. It means explaining what happens, which constraints exist and why the decision matters, while keeping observations separate from assumptions. It is interpretation as well as implementation.

What an FDE does day to day

There is no universal weekly schedule. One task might start by following an order across several tools; another by understanding why someone cannot find a piece of content. Before proposing a feature, it helps to reconstruct the workflow and identify where information gets lost. Then come scope, the people involved and the expected behaviour. The result need not be a lengthy presentation. It should make a concrete decision about the next piece of work possible.

Engineering may involve system connections, data handling, interfaces and automation. A useful first version makes it possible to compare software behaviour with the original need. If there is a gap, it needs understanding before more features are added. My approach is to start with the problem, build the smallest usable thing and improve it by observing its use. I do not confuse the amount of code with the quality of the answer.

FDE vs Solutions Engineer vs Product Engineer

Boundaries vary between companies: this is a comparison for asking questions, not a contractual definition. A title alone does not explain how much work involves development, technical evaluation or customer relationships. To compare roles and responsibilities, I would look at these elements:

  • Forward Deployed Engineer. Starting point: the customer's context. Core work: implement and integrate software usable there. Product connection: explain the needs and limitations discovered, distinguishing customer-specific requirements from those that may be useful more broadly.
  • Solutions Engineer. May focus on evaluating and adopting a solution, demonstrating how it meets a need. Ask whether the role also owns implementation and development or hands the work over after evaluation.
  • Product Engineer. Starts with the product and its use: understands the problem, builds and improves. An FDE may share that mindset while applying it closer to a customer and the particularities of their systems.

Which skills actually matter

Being able to develop software remains essential. So does reading existing systems without assuming they should all be replaced. Incomplete data, integrations and unclear responsibilities may matter more than a framework choice. AI may support development or become part of the solution, but it does not remove the need to understand the workflow. Before integrating a capability, I would ask who checks the result, how it is used and what happens when it is wrong.

Explaining the choices

Communication is not a decorative extra. I taught coding and computational thinking to more than 100 students at Digital Education Lab, so explaining the steps is part of my background. I use AI agents daily and am a Devin AI Ambassador. I do not consider those tools an automatic guarantee of quality: product choices and responsibility for understanding what gets built remain human.

Where this approach appears in my work

For EcoToGo I follow technical work across e-commerce, business software and logistics: TeamSystem Commerce, Danea, order and document transmission, and evaluating a move to Odoo. For BaseCamp Studio I develop semantic search, automatic tags and data enrichment in the DAM. In one case the challenge is information moving between systems; in the other, how an AI capability enters content work. I do not publish unmeasured numerical outcomes or present those engagements under a different historical title.

From a demo to a usable workflow

When assessing an implementation, I would distinguish demonstrating a capability from enabling a person to complete a task. A demo can answer whether a technology can do something. Work in context must answer other questions too: where does the data come from, who can change it, and how does a correction pass through the systems involved? This is where an engineer can translate a local obstacle into an explanation useful to product, without jumping straight to proposing another feature.

An example discussed with the team can describe the original task, the observed behaviour and where the result stops being useful. The response remains open: change configuration, complete an integration or change the product itself. I prefer making that decision explicit rather than calling every problem a bug or every request a priority. Before extending a solution to other customers, understand which constraints belong only to the first context. This is a working principle I would propose, not a claim about a process already adopted at a particular company.

Forward deployed engineer jobs: what to ask

Read a job description beyond its title. I would ask who the customers are, who decides priorities, how much hands-on development is expected and how feedback reaches product. Remote also needs clarification: from which countries, and with which collaboration or presence expectations? I am based in Giovo, Trentino, looking for remote opportunities with AI and product teams across Europe. In-person meetings are possible in Trentino-Alto Adige; other travel is not assumed.

Forward deployed engineer salary: why there is no figure here

I do not have a verified salary benchmark to publish. Employment salary and a freelance engagement fee are not the same thing. Comparing a figure requires the country, arrangement, responsibilities and offer conditions. I do not turn my Fractional CTO retainer into an FDE salary estimate or invent ranges. On contract versus employment, my answer is: let's talk, starting with the role and mutual expectations. Before comparing offers, ask which responsibilities the figure covers and whether the arrangement matches the actual work expected, not only the title.

If you are looking for this kind of engineer, read my FDE profile and the practical cases. I would rather start with a real product or customer problem than a list of buzzwords.

Let's talk

Tell me what's blocking you.

30 minutes, free. Afterwards I'll send you, in writing, the first 3 things I'd do. Or email me: the person doing the work reads and replies.