Bookmarks

Back in my time at @stripe, we had this thing called #crazyideas. It was one of the best cultural mechanisms I've ever seen inside a company. New blog post: https://t.co/38RCMTbkdy https://t.co/DlQJpnpKrc

Coming in OpenCode Desktop v2 10x faster performance (measured) I’m not kidding! I used a 30x cpu slowdown for my baseline at 4fps on an already weak machine, getting to approximately 45fps. - Migrated to tanstack virtual - Removed flicker, scroll (unintentional) jumping...

https://t.co/E9VWMh6TNq

Why does it matter if sandboxes start up in 0.0001 nanoseconds if I then spend the next 20s downloading the git repo?

Did I just unlock claude-fable-5-lite? 😂 Since Fable 5 got pulled (US export control order, Anthropic is contesting it), I wanted to see how much of its character lives in the system prompt vs. the model itself. I ran the leaked Fable 5 prompt on Opus 4.8 head-to-head again...

one way to think on loops is that your agent should bring context to you, rather than you bringing context to your agent

it’s a great day for the “unapologetically celebrate capitalism and progress” section of my book collection https://t.co/iGfWTemzYI

https://t.co/6nWAu8JVdw

For older adults, a daily multivitamin is a high-priority basic, not an optional supplement Across three COSMOS cognitive studies of adults 65+, a standard multivitamin (Centrum Silver) slowed global brain aging by about 2 years The biggest effect showed up in episodic memor...

https://t.co/MVcIdCly31

https://t.co/MCAGg8XjMV

Here's a simple loop: Tell codex to maintain your repos, wake up every 5 minutes and direct work to threads. That makes it easy to parallelize+steer work as needed. I use a orchestrator skill combined with my triage+autoreview+computer use skills, so some work can land autono...

Daniel Kahneman - the psychologist who won a Nobel in economics - spent his life proving one thing: your confidence is lying to you A bat and a ball cost $1.10. The bat costs $1 more than the ball. The answer "10 cents" jumps to mind instantly. It's wrong (it's 5 cents) - and...

Nassim Taleb sat down with Daniel Kahneman - two of the sharpest minds on risk ever - and the takeaway was blunt: stop trying to be smart Kahneman's prospect theory explains why almost nobody can do what Taleb does We're wired to hate the steady trickle of small losses his s...

code mode has helped @cloudflare save ~93% on token cost talking to peers in the industry whose spend is creeping towards upwards of $100k per engineer per year you do the math on the cost savings of that for a large engineering org

BREAKING: Anthropic just dropped Claude Fable 5—this is Mythos, made safe for public release. It is the best coding model in the world. We've been testing it internally @every for the last week or so across coding, writing, marketing, editing, and more—here's our vibe check:...

in the past few days the Cloudflare MCP server made 2.6M API requests from 735k agent tool calls. code mode enables agents to do more per tool call.

You can do this directly in the Codex App: 1. Automations to run autonomous discovery and triage 2. Worktrees to work on isolated features 3. Skills to codify your project/ dev specific knowledge 4. Plugins/ Connectors to connect to tools that you work with 5. Sub-agents to i...

Here’s your monthly reminder that you shouldn’t be prompting coding agents anymore. You should be designing loops that prompt your agents.

the problem with React's useEffect https://t.co/DQ7JALF4KW

During the dot com melt-up of 1995-2000, Nasdaq put up gains of 572%. To date, from the 10/13/22 bear market lows, Nasdaq is up just 162%. Listening to bears calling this a bubble top could cost you a lot of money.

I know people in sales can relate to this one... https://t.co/epSiF6euPo

It’s a beautiful day to learn about AI inference engineering https://t.co/uA4kk0Ople

i’m increasingly convinced that the best agent evals will come from mining real agent failure traces. my view is that every failed trace contains a potential eval but not in its raw form. raw traces are messy, long and too specific. the research problem is to distill them into...

On quality: My first day at @baseten, the very first person I met was @philipkiely on the AI Education team. He told me he's working on a textbook. An absurd thing to say on the surface. I asked for an early pdf version that I started to read on the plane ride home. It was re...

How to reduce Claude Code costs by 70% Add `"showClearContextOnPlanAccept": true` to `.claude/settings.json`. When approving Plan Mode, an option to clear all context except the plan will appear at the top, making it easy to naturally avoid long contexts and reduce token usage.

Insane idea from Opus 4.8 https://t.co/om0CNsO2do

https://t.co/0DyFIXJueI

a little "code mode: react edition" preview! generative ui for any api coming to executor soon

Agents answer questions but throw away all your UI. @RhysSullivan of Executor demoed the fix at MCP Night: Agent Mode — 𝗿𝗲𝗻𝗱𝗲𝗿 𝗨𝗜 lets an agent compose your actual product components and deep-link users back into the product. https://t.co/Fn8xXuL2A8