TOUGH TONGUE AI
  • Home
    LogoTOUGH TONGUE AI

    RESOURCES

    Loading...

    HomeLibraryGen AI Mock · Fine-Tune or Retrieve?

    Gen AI Mock · Fine-Tune or Retrieve?

    Gen AI Mock · Fine-Tune or Retrieve?

    Overview

    A technique-choice round with Felix: prompt, retrieval or fine-tune for a made-up insurer, taken deep. Defend your diagnosis, handle a tuned model that leaks a customer's name, then make a call on your own plan. Written report after.

    Your Instructions

    Your goal: work out what is missing before picking a tool, choose the lightest step that clears a bar, and make a call when the options trade off.

    How this works (about 30 minutes)

    • Felix is an interviewer at an AI company. He starts by asking which role you are interviewing for, and pitches his follow-ups to it.
    • One question, taken deep, then a curveball and a trade-off built from your own plan. Ask for any facts you want; he hands over what he has, and it is made up for the exercise.
    • If you want to sketch or jot, say "can I use the notepad?" and Felix puts it up.
    • No hints and no scores during the call. Your report comes after.

    What scores well

    • Asking what is wrong, knowledge or behaviour, before choosing
    • Treating the two jobs differently, and the lightest step first
    • A test set and a bar made before anything changes
    • Data privacy before tuning, and a real response when it goes wrong
    • A call in the trade-off turn, with a line that would reverse it

    Tips

    • Take Gen AI Coach · Training and Tuning first.
    • Say what you would try first and what would make you step up.

    More in Gen AI Interview Prep: AI Coach Edition

    • Gen AI Diagnostic
    • Gen AI Coach · The Gen AI Loop
    • Gen AI Mock · Have You Worked with AI?
    • Gen AI Coach · How LLMs Work
    • Gen AI Mock · Transformers and LLMs
    • Gen AI Coach · Training and Tuning
    • Gen AI Coach · Decoding and Context
    • Gen AI Mock · The Context Window
    • Gen AI Coach · Prompt Engineering
    • Gen AI Mock · Prompt Engineering
    • Gen AI Coach · RAG
    • Gen AI Mock · Semantic Search for a Chat Assistant
    View the full course (40 sessions) →