Tell me about a pattern you noticed across several customers that changed what your team built.

How to answer

This is the question about the FDE’s second job: carrying what you learn in the field back into the product. Some postings write it down. Google’s GenAI FDE posting asks FDEs to turn repeatable field patterns into reusable modules or product feature requests. ^[1]Source 1Forward Deployed Engineer III, Generative AI, Google Cloud — Google CareersPublisherGoogleSource typecompany job posting Cursor’s FDE postings ask FDEs to turn what they build into reusable patterns and push improvements back into the core product. ^[2]Source 2Forward Deployed EngineerPublisherCursor (Ashby job board)Source typecompany job board A strong answer shows the whole path, from first noticing to the product change.

1. The pattern in one sentence. What several customers needed, and what each one was doing about it: the same script copied between deployments, the same workaround spreadsheet, the same ticket.
2. The moment you noticed. Concrete: “The third customer that month exported our output to a spreadsheet to do the same join by hand.”
3. The evidence, before the argument. A count across all customers, not only yours. What the workaround cost in hours, tickets or incidents. And where the pattern stops: the customers it doesn’t fit.
4. The path onto the roadmap. Who decided, what format they decide in, and the smallest product change that covers the most cases. Say what you offered to do: the spec, the first version, the migration.
5. The result, and your part in it. What shipped, what it replaced, and a number that moved.

Say where the pattern ends out loud. It shows you counted. Colin Jarvis, OpenAI’s head of forward deployed engineering, said in a November 2025 Altimeter interview that the biggest mistake OpenAI’s FDE team made that year was generalizing too early. ^[3]Source 3Colin Jarvis | Head of Forward Deployed Engineering at OpenAI: Trust. Product. Impact.PublisherAltimeter Capital (YouTube)Source typerecorded talk or interview An interviewer who has seen a product bent around two customers will listen for that.

The trap is a story with no path: you noticed, you told product, it happened. Without the evidence and the proposal, the interviewer cannot tell your judgment from luck.