token burn fridays

"treat claude like your flat-earther friend"

I will tell my kids that I lived in the time of load bearing seams

dopamine machine

tldraw roadmap leaked

lmgtfy is now lmamafy (let me ask my agent for you)

my typing has become very sloppy with AI being who I primarily chat to now

I just let Claude rip on a new project, commit straight to main, no CI, and boy do I love what I see

"20 agents are already running, which is the session’s limit"

are there any models for classifying music/sound? or at least lifting out a "feel" or "vibe" of a song in certain categories/classes/flavors

I'm looking for my next role 👀

TS SDKs, DX, agent tooling. remote/hybrid from Boston.

if you've enjoyed ArkType, Standard Schema or my TypeScript perf work, a repost would mean a ton 🙏

ssalbdivad.com

built an extension for this
github.com/frolic/github-…

it's in the chrome store now
chromewebstore.google.com/detail/github-…

funny thing is I didn't get a notification from google about this being approved, but rather a bunch of spam asking me to buy fake reviews

it's in the chrome store now
chromewebstore.google.com/detail/github-…

ai psychosis

trying to figure out how to best distill changes/PRs into something that is easy for me to understand and give feedback on or sign off

at different zoom levels: architecture, APIs/naming, trap doors, trade-offs/alternatives, blast radius, intent/goal

also trying to figure out how my agents can continually zoom in/out to optimize locally and globally (make simpler, more concise, more performant, easier to understand, easier to refactor/delete, etc.)

they often get stuck too zoomed in

I've reached "CI off for PRs" level of AI

opus 5.5 🤝 claude projects

talking about http cookies with @anikrisn
"can I eat the cookie?"
"you can take a byte out of it"
😬