Prince mendiratta

Interested in shipping production grade products in less than a WEEK, no need to spend hours vibe coding? DM!

︎ ︎ ︎ ︎ ︎ ︎ ︎ ︎ ︎ ︎ ︎ ︎ ︎ ︎ ︎ ︎ ︎ ︎

@brexHQ 5) We're a 2-year customer with live operations and a balance to settle. This is the treatment we get. @brexHQ, we want to fix this. Please put a real person on it and give us a reasonable timeline.

@brexHQ 4) Every support contact ends the same way: “I'll escalate this.” No extension. No decision-maker. No real path to fix it. Just a countdown on the account our business depends on.

Imagine running every dollar, card, and payment of your business through @brexHQ for 2 years. Then they give you 7 days to fix your “non-compliant” US address or they close the account. Support's only move is “we'll escalate it.” No extension. No help. 🧵👇

@brexHQ 2) After we submitted an updated address, Brex said they couldn't verify a “compliant physical address” and scheduled our account to close. Our cards, funds, and daily operations all run through it.

@KingBootoshi another good item to add here would be a test loop. every regression found should create a test in the suite to ensure the regression does not happen again.

Sharing another one of my skills I consider S+ tier! I combine this with /goal mode and it just CLEANS shop Agents thrive on positive communication - by making the path to success clear, they will reach the destination! Longer write up on this soon 🫡 github.com/kingbootoshi/d…

Jeff Bezos just revealed how to gain financial freedom: "If you're making $70k or less, try your hardest to start a trillion dollar company." It's that easy folks👏

Why can’t we add GitHub as a provider here????

@threepointone @lgrammel @nicoalbanese10 I too was JUST looking at this yesterday haha

@PriMendiratta @lgrammel @nicoalbanese10 oh damn, thanks!

@lgrammel @nicoalbanese10 do you have a migration doc with breaking/new changes that I can point my coding agent at to see what we should do for agents sdk? please and thank you

Claude Code /goal is insane! Opus spent 5 hours to rewrite the entire system with tests + better architecture! Created more than 800 tests and all of them pass! Truly magical. The product doesn't work anymore though because it hardcoded stuff to make the tests pass but it tried

do not go on TikTok yall 😭😭 someone leaked the entirety of it lol

the king in yellow :

@MonkeyDSpector @xNoRefundsx

@PriMendiratta @xNoRefundsx You not understanding something doesn’t make it vague posting

notice what the kids are not scribbling on

@threepointone ah, amazing! I was under the wrong impression, thanks!

@PriMendiratta it totally runs sandbox sdk, always has, check again

tl;dr - subagent behaviour working on adding multi chat and subagents to the agents starter (yay!) and I have a curious product direction/question. our subagents can be full fledged chats themselves. which means they could not only be async while they work on their thing and you continue, but you could continue "talking" with them after they've "returned" a result. so what should the default behaviour in the starter be? - readonly, no input. this is what most (all?) products/devtools like this do atm - have chat, but it's only followups, doesn't affect the main chat - add a "send back/summarize to main chat" this feels powerful and underexplored I'll probably ship option 1 for now, but there's something here... anyway, multichat/subagents in starter template coming this week

@threepointone last i checked (couple months back), vibesdk works by building a filesystem wrapper over the DO SQLite. It doesn't support UNIX bash commands / a real linux env. I'd like to see an implementation that wraps around sandbox SDK as a real filesystem + linux env access

@PriMendiratta github.com/cloudflare/vib…

@thdxr say WHAT?! No way!

just got inside info that openai is working on a new model