Yi

founder of Humanbased (prev: Codatta, building tools ( for our team, PhD, ex-(Pinterest, Alipay)

[Yi @drtwo101]

Appreciate @MessariCrypto recognizes @codatta_io positioning with deAI data stacks. fun fact: Codatta actually evolves from OSS Microscope Protocol, which we started with @coinbase @MessariCrypto and @GoPlusSecurity https://t.co/uCHCjeTIaC

[Messari @MessariCrypto]

State of AI 2025 🧵

[Y Combinator @ycombinator]

It’s never been easier to design your dream house.

Draw a shape. Define your rooms. Set your constraints.

@DraftedAI generates complete floor plans, elevations, and 3D home designs in seconds.

Over the last month, 120,000 people generated 325,000+ home designs with https://t.co/XqC0LP5n3y.

[Josef Chen @josefchen]

Launching our new paper on arXiv: we trained the largest multilingual food model ever built.

4.1M recipes. 7 languages. 1,790 ingredients. 300 dimensions.

All of human cooking compressed into 2 megabytes.

[Charly Wargnier @DataChaz]

Obsidian users after reading this:

[Shaig @shaigexp]

Fora’s site is live now. Really appreciate the template, @stfnco

Check it https://t.co/qyjPbzhMdA

[Charly Wargnier @DataChaz]

this guy literally breaks down the exact Claude + Obsidian playbook to build your own AI employee in a single afternoon 🤯

[Tom Dörr @tom_doerr]

Compounds knowledge in Obsidian via Claude agents

https://t.co/VgWTASUUch

[Jina AI @JinaAI_]

jina-embeddings-v5-omni is here! Our first universal embedding model for text, images, audio, and video. Available in two sizes: small (1.57B, 1024-dim, 32K context) and nano (0.95B, 768-dim, 8K context). Both support Matryoshka truncation down to 32 dimensions.

v5-omni is back-compatible: if you already use jina-embeddings-v5-text-small/nano, the existing text indexes work with v5-omni out of the box. Without reindexing the text, just index your multimodal content with v5-omni and start searching images, audio, and video.

[Obsidian @obsdmd]

https://t.co/6mKCfJLmao

[Lisa Demchenko @llsbetdigital]

These are some of the flows the app is based on.

They were created by Claude via Figma's MCP and turned out super accurate, outlining not only user flows, but Information architecture and the data used at every step.

This alone is a great foundation for the future app - paired with Design.md, and could potentially replace the prompts that I used.

[rebekah bek @paperseasons]

just launched: @mobbin mcp 💛

your AI agents can now search 600,000+ real app screens. paywalls, onboarding, checkout, permissions — from apps that already shipped.

ai tools can write code. they just don't know what good looks like. now they do.

👉 https://t.co/WL3IofRU84

[Yi @drtwo101]

@dingyi 刚release一个最新的稳定版本优化了很多体验和稳定上的问题，0.9.0

[Yi @drtwo101]

@dingyi 欢迎反馈，这个是我们内部build的小工具，正在向openai的symphony迁移，不过这个auto cr pipeline对于习惯和coding agents直接打交到的同学还是比较舒服。有什么问题可以欢迎提issue.

[Yi @drtwo101]

@dingyi

[Yi @drtwo101]

@dingyi 欢迎使用 https://t.co/8QZ6FVHduO 自动化Claude Code build -> Codex review

[Yi @drtwo101]

@hannah_hearth implement
"Make Claude and Codex do Pair Programming" (mentioned in Vercel team's best practice) via https://t.co/8QZ6FVHduO (`npm install @motivation-labs/crosscheck@beta` for lastest update). The crosscheck works like below

[Yi @drtwo101]

@hannah_hearth implement
"Make Claude and Codex do Pair Programming" (mentioned in Vercel team's best practice) via https://t.co/8QZ6FVHduO (`npm install @motivation-labs/crosscheck@beta` for lastest update). The crosscheck works like below

[Goodfire @GoodfireAI]

Neural networks might speak English, but they think in shapes.

Understanding their rich *neural geometry* is key to understanding how they work – and to debugging and controlling them with precision.

Starting today, we’re releasing a series of posts on this research agenda. 🧵

[Yi @drtwo101]

Claude writes code. Codex reviews it.
Codex writes code. Claude reviews it.

That’s the loop behind crosscheck — a local, cross-vendor AI code review orchestrator for PRs.

Built for catching agent mistakes before they land.

https://t.co/KUG5P6CGkD