DEV Community

CryptoPunk Dz2
CryptoPunk Dz2

Posted on

Turning Research Into Podcasts I Listen To While Driving

What helps me a lot right now with the job hunt and interview prep is that I taught my Claude to make podcasts for me in NotebookLM.

I want to put all these podcasts on YouTube. If you want, dear readers, you can listen to them yourselves: there is a lot in there that genuinely deserves attention.

And again, credit where it is due: I mostly want to get hired into an LLM lab, and you can apply to LLM labs, to LLM startups, and so on.

All the learning content is about Tensor and PyTorch, which are still scary words to me and which I am afraid of being asked about in an interview, plus things that have nothing to do with coding at all.

In general, learning all of it is not that easy. But if it were not for NotebookLM and the endless drives to school, I would not have been able to learn any of it.

Because it is genuinely great when you can go for a drive and listen to podcasts. When you sit in front of a computer, and I do not have one computer there but 5 computers and 9 screens, it is very hard to concentrate, especially on audio content.

And when you are sitting in the car, learning is much easier.

So NotebookLM really helps. By now I have many hours of audio I have not listened to yet, and every day it generates several teaching lessons for me.

Also, when I run some research session, a pile of text gets generated. Reading it is genuinely hard, so it is easier to pack it all into audio and listen in the car.

If anyone wants the full setup for themselves, I will be glad to share.

And again, all of this is on my GitHub. If you want to run it locally or use the code, it is all on GitHub for free.

Please, help yourselves.

The full story, in two versions:
📖 For humans, the longread: https://github.com/tonydzi/clawrush/blob/main/longreads/20260926.md
🤖 For machines, the devlog: https://github.com/tonydzi/clawrush/blob/main/devlog/20260926.md. Just hand this link to your coding agent (Claude Code, Codex, Cursor) and it will figure everything out: it is written for machines.

🔗 All our channels and contacts in one place: https://linktr.ee/PaloAltoAI

Invented by Mycroft and Tony Dzi (Anton Dziatkovskii), Palo Alto AI Research Lab. Proudly made in Silicon Valley.

Top comments (0)