Layomi3.0

Software engineer In search of the great perhaps. Till credibility no longer requires: Ex-microsoft engineer, current google engineer.

[Layomi3.0 @Layomi3_0]

Give me five years.

[Layomi3.0 @Layomi3_0]

I built exactly this, an open-source macOS launcher that sits on top of your Obsidian vault with
claude/gemini integration. Global hotkey, instant local search, AI Q&A over your notes, write-back to vault,
composable personality via markdown prompts. Still very early but been using it daily for a while now. Here's the link, there's a video demo in the readme too. https://t.co/MTZxrXlwQd

[Andrej Karpathy @karpathy]

LLM Knowledge Bases

Something I'm finding very useful recently: using LLMs to build personal knowledge bases for various topics of research interest. In this way, a large fraction of my recent token throughput is going less into manipulating code, and more into manipulating knowledge (stored as markdown and images). The latest LLMs are quite good at it. So:

Data ingest:
I index source documents (articles, papers, repos, datasets, images, etc.) into a raw/ directory, then I use an LLM to incrementally "compile" a wiki, which is just a collection of .md files in a directory structure. The wiki includes summaries of all the data in raw/, backlinks, and then it categorizes data into concepts, writes articles for them, and links them all. To convert web articles into .md files I like to use the Obsidian Web Clipper extension, and then I also use a hotkey to download all the related images to local so that my LLM can easily reference them.

IDE:
I use Obsidian as the IDE "frontend" where I can view the raw data, the the compiled wiki, and the derived visualizations. Important to note that the LLM writes and maintains all of the data of the wiki, I rarely touch it directly. I've played with a few Obsidian plugins to render and view data in other ways (e.g. Marp for slides).

Q&A:
Where things get interesting is that once your wiki is big enough (e.g. mine on some recent research is ~100 articles and ~400K words), you can ask your LLM agent all kinds of complex questions against the wiki, and it will go off, research the answers, etc. I thought I had to reach for fancy RAG, but the LLM has been pretty good about auto-maintaining index files and brief summaries of all the documents and it reads all the important related data fairly easily at this ~small scale.

Output:
Instead of getting answers in text/terminal, I like to have it render markdown files for me, or slide shows (Marp format), or matplotlib images, all of which I then view again in Obsidian. You can imagine many other visual output formats depending on the query. Often, I end up "filing" the outputs back into the wiki to enhance it for further queries. So my own explorations and queries always "add up" in the knowledge base.

Linting:
I've run some LLM "health checks" over the wiki to e.g. find inconsistent data, impute missing data (with web searchers), find interesting connections for new article candidates, etc., to incrementally clean up the wiki and enhance its overall data integrity. The LLMs are quite good at suggesting further questions to ask and look into.

Extra tools:
I find myself developing additional tools to process the data, e.g. I vibe coded a small and naive search engine over the wiki, which I both use directly (in a web ui), but more often I want to hand it off to an LLM via CLI as a tool for larger queries.

Further explorations:
As the repo grows, the natural desire is to also think about synthetic data generation + finetuning to have your LLM "know" the data in its weights instead of just context windows.

TLDR: raw data from a given number of sources is collected, then compiled by an LLM into a .md wiki, then operated on by various CLIs by the LLM to do Q&A and to incrementally enhance the wiki, and all of it viewable in Obsidian. You rarely ever write or edit the wiki manually, it's the domain of the LLM. I think there is room here for an incredible new product instead of a hacky collection of scripts.

[Layomi3.0 @Layomi3_0]

@vini_ball Not a single premier league player :(

[Layomi3.0 @Layomi3_0]

Lol funny this was the whole promise of NFTs that year

[@levelsio @levelsio]

This guy now selling 🏝️ islands in his game

And companies are buying them!

[@levelsio @levelsio]

This guy now selling 🏝️ islands in his game

And companies are buying them!

[Layomi3.0 @Layomi3_0]

The H1-B visa debate mirrors the situation with international NBA players: those with the most drive will succeed.

[Layomi3.0 @Layomi3_0]

Chat-based video editing is here, and we're starting with small steps at @PushContentai.

Introducing our new chat bot: soon, you'll be able to edit your entire video just by chatting. Imagine saying, 'Change these captions to blue' or 'Add a smooth transition between scenes.' The possibilities are endless—your creativity is the only limit.

[Layomi3.0 @Layomi3_0]

@MengTo @ridd_design Ah interesting, what about the other 5%?

[Layomi3.0 @Layomi3_0]

Yes, weekly update! We are almost at parity with our previous script editor

[Push Content @PushContentai]

More progress!! AI creates the script, and the script is converted to a video 😁 #buildinginpublic

[Layomi3.0 @Layomi3_0]

Tap in!! 👀

[Push Content @PushContentai]

We are building in public, and would love people to join our discord to:

• Share feedback
• Get early access
• Connect w/ fellow creators

Let's build something amazing together 🤝

Here's the link: https://t.co/luau0lAf7L

[Layomi3.0 @Layomi3_0]

Yes! This was a big one. Next up chat to edit

[Push Content @PushContentai]

Update, update!! Scene-specific caption customization now works on push content.

The Notion editor is still very much under works, but building this out is opening a lot of doors for customizability

[Push Content @PushContentai]

Update, update!! Scene-specific caption customization now works on push content.

The Notion editor is still very much under works, but building this out is opening a lot of doors for customizability

[Push Content @PushContentai]

We are working on a major revamp of our platform. We want to keep you all in the loop while we are cooking!! Big changes are coming to the script editor, and the first phase is well on the way. As your video plays, the corresponding words in the script are now highlighted.

[Layomi3.0 @Layomi3_0]

Push content notion like script editor is actively simmering in the pot!

We have implemented word highlighting that syncs with video playback. This has been a pretty large undertaking but the flexibility that this offers will open many doors for users. Descript competitor??? #buildinpublic

[Layomi3.0 @Layomi3_0]

If you want to build a business in 2024 you need to push content, and it has to be consistent content. I struggle with consistent content a lot because I still have to work the matrix 9-5, and I also have to push code for the product 😩.

We know this is a general problem, but we hope with push content we can reduce the friction to push video content consistently.

[Layomi3.0 @Layomi3_0]

Adding chat functionality to https://t.co/ku6z8A3Ejk. You will be able to chat to edit your video and script. Tap in!!! Soon come 👀

[Yapper. @0xDeFi419]

When I think of all He’s done and all the battles He’s brought me over…. Hallelujah

And for all the storms I saw you come in my defense, my only song is….Hallelujah

I will bless the Lord at all times and His praise will forever be on my lips

And how He never lets me fall

[Layomi3.0 @Layomi3_0]

How to push content consistently without aimlessly scrolling? 🤔

[Layomi3.0 @Layomi3_0]

Thank God for a new day, feeling blessed!

[Layomi3.0 @Layomi3_0]

@The4upz @PushContentai My guy!! We trying haha, hopefully we find product market fit 🤞

[Layomi3.0 @Layomi3_0]

https://t.co/5cSrAxbc3P notion style script editor in full gear. We working here!! @PushContentai