Bookmarks

Cursor put out a blog today stating that semantic search beats grep Semantic search doesn't have to be complicated, and thats exactly why I built SemTools -- to provide agents with a "fuzzy semantic grep search" Semtools https://t.co/Zih7K4skqm Blog https://t.co/pc9cFAd4ln

Here’s the part where he was talking about @tszzl Great convo https://t.co/OnOEgleBSo

it's just so easy to build a digital garden with Claude/Codex https://t.co/GVFk5CL6qa

What's your favorite Claude Skill?

Not VC-funded. Solo. Not in SF, in Bulgaria. Introducing Notate — made for Notion. Because note-taking isn’t the goal anymore. Effortless context generation is. https://t.co/D7ppxXcjmQ

My excellent Conversation with Sam Altman: https://t.co/YkUAJnglJm, @sama

just read @Alex_Danco's blog on jevons and baumol paradoxes and it perfectly explains the strange economics we're about to see with AI here are the 3 insights that stood out: 1. jevons paradox - when things get cheaper, we use MORE of them, not less a transistor cost $1 in ...

New on the Anthropic Engineering blog: tips on how to build more efficient agents that handle more tools while using fewer tokens. Code execution with the Model Context Protocol (MCP): https://t.co/PeStmufIkp

Ran into so many troubles with Plasmo, switched to WXT for the Chrome Extension. Much better tooling. Works far better in a monorepo. https://t.co/Xg4iZl6rBM

Bookmarked. Can’t wait for when I can finally sit down to read this

Join our Building with Cursor event livestream from our NYC HQ! Learn how designers from @cursor_ai, @NotionHQ, and Ramp are changing their workflows. Hosted by @lalizlabeth https://t.co/QNp48wbjI0

Our tech headcount strategy in a nutshell 1. Hire any engineer that clears our (extremely high) bar 2. Maintain ratios with xfunctional team ~1:5 design, ~1:10 PM, ~1:10 DS, ~1:30 PMM, ~1:20 POPs 3. Cap growth rate (<50% YoY) to maintain culture, onboarding productivity, and...

I totally stole this and converted it into a Claude Skill. I have already stopped using browser MCPs before, but did not find a good replacement. This one does not look completely terrible. https://t.co/92AOSaYSPl https://t.co/e8eaY675rg

https://t.co/1PgCbLsAtb

New blog post, wherein I beat a dead horse for the last time. https://t.co/j64CpIysH1

An M.I.T. study found that 95% of companies that had invested in A.I. tools were seeing zero return. It jibes with the emerging idea that generative A.I., “in its current incarnation, simply isn’t all it’s been cracked up to be,” @JohnCassidy writes. https://t.co/DD2C4FXfgz

Peter compares ChatGPT (GPT-5), Claude (4.5), and Gemini (2.5) across 10 tasks, recommending the best model for each based on personal use. - Prefers ChatGPT for everyday answers due to speed, conciseness, and fewer rate limits compared to Claude. - Chooses Claude for writing...

BEST SMALL AND OSS AI MODELS open source - deepseek v3.2 fine tune base- qwen family overall small model - Flash 2.5 agentic - GLM, Haiku 4.5 coding - grok-4 code, qwen coder classifier -gpt-5 mini reasoning - Deepseek R2

Get these and thank me later https://t.co/MsMV4Eg8Z8

this tool for breaking down amazon boxes is the most life-changing purchase under $20 i’ve ever made h/t @lennysan for the rec https://t.co/ND1tX7rsd4

Whenever I feel lost, I watch this. https://t.co/RN5EJK2BB6

I still think about this a lot https://t.co/v9vcV4Zq8d

🚨New Anonymous Model: Cedar Available on designarena. My best guess, not sure: GPT 5.1. It's rare to encounter, so limited testing. Here's a Discord clone it created, like other GPT models, it reimagines the app in GPT 5's distinctive style rather than copying. Code belo...

This is a fascinating way of working with multiple coding agents at once: "Before doing anything else, read ALL of AGENTS dot md and register with agent mail and introduce yourself to the other agents. Then coordinate on the remaining tasks"

If you're setting up your Coding CLI, here are some tools worth adding. I've found fd, ripgrep, and ast-grep especially useful when working with codex. https://t.co/qS6HcBNiVq

My coding agent workflow has really changed a lot ever since I gave them access to messaging so that they can directly communicate with each other. Now, I have one of them come up with a super detailed plan and sometimes have GPT Pro review and improve the plan in the webapp. ...

BREAKING: John and Jordi join Ilya and Marc on today’s show https://t.co/tjeAOt8QvX

Some good tips in there, also A LOT of workarounds to fight Sonnet's post-training choice for rushing instead of being diligent and careful. With codex you can drop 90% of the charade. https://t.co/IuRx9AX8wR

Hello Thermo World. https://t.co/yO96FRhoZ6

Updated my AI model capabilities table. Here's my favorite model for each task: 1. Everyday answers: I use ChatGPT because GPT-5 is fast, concise, and almost never hits rate limits. 2. Writing: Claude is my main editor. I give it examples of my best writing and it just gets...