I Used Gemini Notebook and Claude to Interview Me for the Job
3 Interviewer Personas, Quiz, Mind Map, And One Real Job Posting.
A couple of days ago, I connected Gemini Notebook to Claude to build a study system that only quizzes you on your own sources. In that post, I mentioned that there’s a 2nd part to it. Well, here it is.
This post is more about preparing, rather than learning. Or...it can be both.
We will be using the same tools, but in a different way.
This time, we are preparing for an interview. It can be something that you’ve been preparing for, and want to take the lest preparation step. It also can be your dream job and you want to find a roadmap towards it.
The thinking and the folder system can be applied to any other field. You can apply this same thinking to a marketing job, sales, etc. The same tech lead and hiring manager roles will run against that instead. The one role I tested is technical enough that it splits pretty well into technical questions (tech lead) and workflow/experience questions (hiring manager). You can always modify the personas to fit your needs.
By the end of this post, you’ll have:
A working pipeline that turns a real job posting into a grounded interview quiz and mind map
Three distinct interviewer personas (hiring manager, tech lead, CEO), each pulling from a different part of whatever source you provide
The actual folder to download and point at your own target job
Here’s what’s actually inside that folder:
Let’s build it!
The mechanism & what you point it at
Everything from last post still applies:
You use “notebooklm-py” that bridges Gemini Notebook to Claude Code
You create a quiz or mind map from your source
You find whatever job you want to include
I also should note here that nothing in the “interview-panel“ folder is specific to NVIDIA, or to engineering roles. The router and the personas only point at whatever “quiz.md“ and “mindmap.json“ your own notebook generates.
Whenever I thought about this post, I, for some reason, immediately thought about NVIDIA. Maybe, because I saw that Jensen posted on X for the first time:
I didn’t go into the specifics of the job for this post, but I just searched for an “AI Engineer” position and took the first one I saw.
Another disclaimer here:
The personas in this post are built from the pattern of what this specific posting asks for. They’re not a claim about how NVIDIA runs its interviews. I don’t know how they do it exactly.
The only thing I know about that job is just the job posting and that’s it.
Below the paywall:
The exact commands, step-by-step instructions, a couple of questions from the tech lead and hiring manager, and the “interview-panel” folder itself to download and point at your own dream job.
