An FDE engagement is not "here is a ticket, go build it." It is a relationship that moves through phases: discovery, scoping, pilot, deployment, and adoption. Knowing where you are in that arc tells you what to optimize for right now.
Why this appears in interviews
Interviewers ask behavioural questions like "walk me through a complex project you owned end to end." They want to hear that you understand the whole arc — not just the coding, but the discovery before it and the adoption after it. Candidates who jump straight to "then I wrote the code" reveal they have only ever lived in the middle of the lifecycle.
The phases
Discovery. You learn the customer's actual problem, their systems, their constraints, and who the stakeholders are. The goal is to separate what they asked for from what they need. Most failed engagements were lost here, by building the literal request.
Scoping. You turn an ambiguous problem into a small, shippable first slice. The art is picking a slice that delivers visible value fast and de-risks the unknowns, rather than boiling the ocean.
Pilot. You build and deploy the first slice in their environment, with real (often partial) data. You are proving the approach and earning trust, not shipping the final product.
Deployment. You harden the pilot for production: error handling, security, monitoring, and the integration edges you deferred. This is where pilot purgatory is won or lost.
Adoption. The system is live; now people have to actually use it. You train their team, watch real usage, fix what confuses users, and hand off so it runs without you.
The two-way street
A great FDE is also a conduit back to the product team. The friction you see in the field — the missing feature, the confusing default — is the most valuable product feedback the company gets. Bringing that back is part of the job, not a side effect.
Every phase has a different definition of "done." Discovery is done when you understand the real need. Pilot is done when the customer believes. Adoption is done when they no longer need you.
Common interview mistakes
Skipping discovery. If your project story starts at "I built X," add the part where you figured out X was the right thing to build.
Declaring victory at launch. Launch is not adoption. Mention how you confirmed people actually used the system and what you changed based on real usage.
Key vocabulary
- Scoping — turning an ambiguous problem into a concrete, shippable first slice.
- Pilot — a deployed proof-of-value with real data, ahead of full production.
- Adoption — the phase where success is measured in real usage, not deployment.