Minchy Cheng

Just a normal person who trying to survive in the AI world

[OpenAI @OpenAI]

Codex now works directly in Chrome on macOS and Windows.

It’s even better at working with apps and sites in Chrome, and now works in parallel across tabs in the background without taking over your browser.

To get started, install the Chrome plugin in the Codex app.

[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.

[Stitch by Google @stitchbygoogle]

Meet the new Stitch, your vibe design partner.

Here are 5 major upgrades to help you create, iterate and collaborate:
🎨 AI-Native Canvas
🧠 Smarter Design Agent
🎙️ Voice
⚡️ Instant Prototypes
📐 Design Systems and DESIGN.md

Rolling out now. Details and product walkthrough video in 🧵

[Ahmed Omar. @omar_or_ahmed]

Every healthcare AI co is building for clinicians.

Nobody's building for patients.

That's the $50B mistake.

[UBI Works 🇨🇦 @ubi_works]

4,000 jobs cut out of 10,000 directly attributed to AI. Stock price soared. UBI is inevitable.

[Ejaaz @cryptopunk7213]

fucking hell this confirms my worst fear

karpathy estimated the risk of AI automating your job and er… its not looking great - 42% of jobs are at high risk (60 million people)

if your job uses a computer you might want to look at this:

- 42% of U.S. jobs score 7+ on exposure, that’s ~$3.7 TRILLION in annual wages.

- most at-risk jobs: medical transcriptionists (10/10), computer programmers (8/10), financial analysts

- safest jobs = plumbers electricians and even nurses.

takeaway: if your job involves your hands you’re good (for now), ai needs a human body for physical labor

but if your job is on a computer. cooked.

anthropic’s report last week confirmed the same data.

U.S. bureau of labor statistics ALSO predicts this.

elon yesterday: “all jobs will be optional”

[Minchy Cheng @MCheng22202]

@hsn8086 头像越萌实力越强

[Claude @claudeai]

1 million context window: Now generally available for Claude Opus 4.6 and Claude Sonnet 4.6.

[AI Will @FinanceYF5]

🚨 Anthropic CEO 警告：公司已无法确定 Claude 是否没有意识。

[Minchy Cheng @MCheng22202]

@PANews 腾讯现在连外国人也要坑吗 笑死

[Minchy Cheng @MCheng22202]

@KKaWSB 小哥的提示词注入用得挺溜的

[Elon Musk @elonmusk]

@Polymarket Just issue checks from the Treasury.

When good & services output growth due to AI/robotics far exceeds growth in the money supply, there will be massive deflation.

[Minchy Cheng @MCheng22202]

@TechFlowPost 套壳公司能做到这种程度也很强了

[TechFlow 深潮｜APP 已上线 @TechFlowPost]

https://t.co/8h9QqFBHtD

[Claude @claudeai]

Claude can now build interactive charts and diagrams, directly in the chat.

Available today in beta on all plans, including free.

Try it out: https://t.co/tHPAZRgQkn

[Minchy Cheng @MCheng22202]

@Kalshi Where does the money come from? Massive money coming in these big AI companies. Are they really that kind to give away their money 💰?

[Alex Finn @AlexFinn]

Everything you need to master in 2026 to get rich:

• Vibe coding
• OpenClaw
• Running your own local models
• Codex
• Karpathy's Autoresearch
• Building an X audience
• Creating videos
• AI agent swarms
• How machine learning works
• How databases work (been using Convex)
• Using API's
• Training your own LoRAs
• Elimination of doom scrolling

[RainbowBird | 洛灵 @luoling8192]

很喜欢《软件设计的哲学》这本书，于是写了个 skill，帮助大家写出质量更高的代码：
https://t.co/VqR4XcVvK1

[Perplexity @perplexity_ai]

The Perplexity API platform is now a full-stack, model-agnostic API platform for building agents.

It replaces your model provider, search layer, and embeddings, built on the same infrastructure that powers Perplexity.

[Andrej Karpathy @karpathy]

Expectation: the age of the IDE is over
Reality: we’re going to need a bigger IDE
(imo).

It just looks very different because humans now move upwards and program at a higher level - the basic unit of interest is not one file but one agent. It’s still programming.