Samuel Wei

An explorer in artificial intelligence; DeepShare Inc Co-Founder

[Samuel Wei @SamuelWei10]

Tell Javis what do you want to remember?

[Samuel Wei @SamuelWei10]

Javis, do you remember anything?
Tell Javis something you’d like it to remember. 👇

[Samuel Wei @SamuelWei10]

@Stanford Congratulations

[Samuel Wei @SamuelWei10]

Javis Earphones as my AI companion.

Capture ideas anytime, anywhereg with a simple voice note. No typing, no interruptions — just speak and let Javis remember.

[Samuel Wei @SamuelWei10]

Meet Javis Earphones — your always-ready AI companion.

Capture ideas anytime, anywhere with a simple voice note. No typing, no interruptions — just speak and let Javis remember.

https://t.co/ROzaiDaXX6

#AI #PersonalAIAgent #SecondBrain

[Samuel Wei @SamuelWei10]

As your reading and listening speed increases, how do you take notes?

Can your note-taking speed still keep up with the rate at which you’re consuming information?

This is a method I’m currently experimenting with. I’d love to hear your critical feedback!

[Samuel Wei @SamuelWei10]

https://t.co/oLJ87XHDmJ Apple Store download

[Samuel Wei @SamuelWei10]

Agent live in your HiJavis customize keyboard

[Samuel Wei @SamuelWei10]

HiJavis v0.4.3 is out 🎧

Using Bluetooth earbuds now feels much more natural — start or stop recording directly from the earbuds, hear subtle private audio cues, and enjoy a cleaner, less intrusive lock-screen experience.

Demo: https://t.co/Y3ukZDYwBo

[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.

[Samuel Wei @SamuelWei10]

#AI #secondbrain #Javis

[Samuel Wei @SamuelWei10]

HiJavis v0.3.3 just dropped—summaries now chunk your recordings neatly and pull out key insights. Voice keyboard proofreads on the fly, keeps your raw audio synced with the sleek text. Revert anytime.
Download on iOS: https://t.co/oLJ87XHDmJ

[Samuel Wei @SamuelWei10]

We dreamed of creating an AI assistant that truly understands us — one that captures our raw, rarely perfect first thoughts and transforms them into clarity. Most ideas fade before they take shape.

[Samuel Wei @SamuelWei10]

https://t.co/oLJ87XHDmJ
Download HiJavis on apple store

[Samuel Wei @SamuelWei10]

Meet HiJavis

[Samuel Wei @SamuelWei10]

Good using with AI [HiJavis](https://t.co/oLJ87XHDmJ) that it’s your biggest leverage.

[Samuel Wei @SamuelWei10]

While you are hard thinking, try to use the HiJavis (download at Apple Store) to record what your thoughts from your voice to the texts.

[Samuel Wei @SamuelWei10]

Meet HiJavis — your second brain for real-time idea capture.
✨ Frictionless.
⚡️ Concise.
📶 Works even offline.
Because your best ideas deserve to be remembered — not lost.
#AI #Productivity #Javis #SecondBrain

[Samuel Wei @SamuelWei10]

Matrix is very cool

[Tivadar Danka @TivadarDanka]

Matrix factorizations are the pinnacle results of linear algebra.

From theory to applications, they are behind many theorems, algorithms, and methods. However, it is easy to get lost in the vast jungle of decompositions.

This is how to make sense of them.

[Samuel Wei @SamuelWei10]

Agents are only functions

[Santiago @svpino]

You should stop treating agents like humans.

This has probably been one of the biggest unlocks for me over the past few months in terms of building agents.

I see people giving agents personalities, goals, and even titles.

But agents aren't people — they are functions with context windows.

When you anthropomorphize agents, you can't see them for what they are.

This might sound weird, but errors didn't make sense because I was thinking about agents the wrong way.

Agents are functions. It's much better to think of them as unreliable APIs that will give you the correct answer most of the time.

As soon as I did that, my code became cleaner and much more efficient.