- give yourself a concrete algorithm practice goal, such as "get through Blind 75" (it's a list of 75 questions you can find online).

- practice using data structures and also implementing whatever your language doesn't provide

For context, I just spent most of the last two months doing a job search in the Bay Area. Did final round interviews at several companies ranging in size from a few people to a few thousand.

I encountered exactly 1 (one!) direct leetcode problem during my interviews.

Of course, it should also be possible to completely disable Gemini so as to avoid accidentally sending it private browsing content.

The LLMs are quite good at writing one-off Obsidian plugins, in my experience.

I think the family plan has great value if you have family members that wouldn't otherwise pay for it. Gifting others with less ads in their lives is a significant win.

> The app will eventually become obsolete. It’s the plain text files I create that are designed to last.

https://cerebras-inference.help.usepylon.com/articles/192554...

There's a demo video in the readme.

Why?

Obsidian is a note taking app with tons of extensions. Even so, there must be hyper-niche use cases that aren't being served by any existing extension. LLMs are decent at coding though, so maybe an LLM can write custom functionality on demand.

That's the experiment, to see if you can customize your notes app simply by asking the LLM.

The obvious caveat is that the plugin is only as good as the LLM driving it. I've been very pleasantly surprised by how capable Sonnet 3.7 is at writing Obsidian code (without examples). However, even so, it is far from perfect.

Is this useful?

It's _interesting_, but I'm not sure yet how useful it is. The AI doesn't always succeed, and the combination of prompt + model will make or break the experience.

I'm looking to get feedback.

Broadly speaking I'd love to see more software that could be customized to meet extremely specific user needs.

[1]: https://obsidian.md/

Comments URL: https://news.ycombinator.com/item?id=43781688

Points: 2

# Comments: 0

Some breaking changes in there. May want to hold off on upgrading until LLMs come around to writing v4 code with ease.

This is not always possible, but if the product in question has a mobile app or a wearable talking to a server, you might be able to utilize the same API it's using:

- intercept requests from the device - find relevant auth headers/cookies/params - use that auth to access the API

- copilot-chat was just a chat sidebar last time I used it, you still had to manually apply any code suggestions. cursor will apply changes for you. It's very helpful to have a diff of what the AI wants to change.

It's been a while since i've used copilot though, so copilot chat might be more advanced then i'm remembering.

edit: formatting