Forward deployed engineer vs solutions engineer

A role may combine discovery, demonstrations, code and production support, while another employer divides those tasks among several teams; do not infer that a solutions engineer cannot code or that every FDE owns long-term operation.

OpenAI’s FDE role description-sf-san-francisco/) provides a concrete example of discovery-to-rollout ownership, while Anthropic’s staff FDE description illustrates customer-embedded engineering.

Use forward deployed engineering to understand that delivery model, then compare the actual role documents and proposed deliverables rather than turn those employer examples into rules for every organisation.

For each activity, ask who performs the work, who approves it and what output remains with the client; include production permissions, repository access and the process for changing configuration.

A forward deployed AI engineer brief can specify model evaluation and retrieval work, while AI integration services identifies system-boundary responsibilities that any delivery role may need to cover.

In the illustrative account-briefing project, someone must own source permissions, unsupported claims and failed CRM updates even if several engineers or vendors contribute; the matrix reveals gaps the titles cannot resolve.

If the task is unclear, prioritise discovery and technical explanation; if the task is agreed but unbuilt, prioritise engineering evidence, acceptance checks and the ability to transfer ownership.

Hire forward deployed engineers compares engagement and hiring considerations, while how to hire forward deployed engineers offers a structured assessment approach.

Ask candidates or partners to explain the same representative workflow, including a failure and a handover, then compare their answers against the responsibility matrix rather than a preferred title; document any work another team must supply.