At that time I was already working as Head of Data, so it was a bit humbling. After that, I changed my approach and managed to pass the next three interviews.

Most people prepare the wrong way for AI Engineer roles. They focus too much on theory (transformers, attention, papers) and not enough on what companies actually test:

  • Can you build and ship AI features?
  • Do you understand RAG, agents, and evaluation?
  • Can you talk about cost, latency, and reliability?
  • Do you have real projects you can explain deeply?

In the full article I broke down:

  • What AI Engineers actually do day-to-day
  • Skills that matter vs skills that don’t
  • Real interview questions (with how to answer them)
  • How companies evaluate LLM applications
  • 3 projects that impress hiring managers
  • A practical 12-week prep plan

Here’s the full write-up:
https://medium.com/the-agentic-architect/how-to-prep-for-first-ai-engineer-interview-and-actually-get-the-offer-in-2027-b1ebe2b9fab0

Would love to hear from others — what’s been the hardest part of AI Engineer interview prep for you?