What Your AI Can’t See

October 30, 2026 – 9:00 am

Your AI coding assistant guesses about your project and never tells you it guessed. Working in Claude Code on a real ML codebase, you will see where that happens and learn the context habits that stop it

Hosted by

Cihan Yatbaz Data Science Educator Xebia Academy
Register now

Details

TARGET AUDIENCEData scientists, ML engineers and analytics engineers who already use an AI assistant on real work and keep finding it unreliable on their own code. Demos run in Claude Code, but the ideas carry over to Copilot, Cursor or Gemini, and no prior Claude experience is needed.LEARNING GOALSBy the end of the session you will be able to:

  1. Say what an AI coding agent can and cannot see in your project, and predict where it will quietly guess.
  2. Write aCLAUDE.md, the project instruction file that tells Claude Code how your repo works, and see the same prompt produce a different answer because of it.
  3. Put any prompt that changes a file through a four point check: scope, criteria, plan, verify.
DURATION60 minutes.REQUIREMENTSA laptop is optional, because everything is demonstrated live in Claude Code. To code along you need VS Code and any AI assistant you already have access to (preferably Claude).

Register now