Vincent van der Meulen

@LuthraAbhyuday yes!

@vinvan one laptop for work that stays in the office?

the 'agentic era' is ruining my health so i started run commuting. it's a really nice, efficient way to exercise! this is attempt 1000 though. what's making it stick this time: 1. running home, not to work 2. two laptops so you don't need a bag ymmv (literally)

@lucasmeijer i'd hack on top of github.com/ColeMurray/bac… or github.com/vercel-labs/op… — they both have great defaults

if I do want to lego my own together, which lego blocks should I look at? what do you like for sandbox? what do you like for agent loops? what do you like for "somehow persist my sandbox state"

Who has a really nice setup like this? - Cloud based agents - Kanban board-esque overview - Full control over agent loop

@theesabina @NYCMayor i refuse to move to brooklyn 😭

@vinvan @NYCMayor he only wants people to prompt on the L between 1st and Bedford (no joke the best signal in the city)

if mamdani was truly the people's mayor he'd make subway prompting easier. it's too loud on the platform to voice dictate, and service is spotty underground. millions of agi pilled nyc'ers are suffering @NYCMayor

This is really neat but it’s not a design tool as much as it’s a design _production_ tool. The practice of design is mostly about what comes before production. There’s no doubt in my mind that all parts of software production will become automated very soon. Writing code, making web pages, putting pieces of a design system together etc. And that’s fine. I think few people actually enjoy this kind of production work. Wouldn’t it be better if we spent our precious time in life on what is more meaningful?! At the core, the practice of design is methodical; like architecture, not like art. In a nutshell: We find constraints, form comprehension of the whole and propose solutions that honor those constraints. First after that do we enter some form of production phase, usually prototypes first, learn about some constraints that were hidden before, loop back, prototype and then build the production-grade “final” artifact. These last few tasks are quickly losing value because AI tools can do it much faster (not yet better though) than humans. It’s simply just what has the best RoI for a business. Some companies and individuals will continue to spend human time on certain parts of the “production line” as a market differentiator, but it will cost them a relatively high price compared to competitors. Anyhow, I still haven’t seen a tool better than Figma that supports the actually-interesting part of the design process. I wouldn’t be surprised if Figma focused their products on that, maybe separating “products for production” of “products for ideation & exploration.” The latter would obviously still leverage AI, but not to do the work for me but rather to support my efforts the way a therapist helps me live a better life (not living my life for me.)

i'm @vinvan now

@jsngr oh wow this explains a lot

AI-pilled

@CompuIves really loving all the eng breakdowns! looks awesome

Almost everything in Biscuit runs in the browser. Code gets compiled using Oxc (wasm), the LSP runs in a web worker, and the agent harness too. This allows for really quick direct manipulation. You can click on any element and edit it without server roundtrips.

managing parallel agents gets tiring... so you should use a 'middle manager' (h/t @sawyerhood) a very easy way is to tell a @cursor_ai cloud agent it can use the cursor cli to spawn other cloud agents!

as models get better, ai design tools should optimize for what happens before and after your ai designer starts designing. i'd love to see more fresh takes on managing inspiration, reviewing, riffing, simulating, testing, and designing async!

give me this and an LLM

@jsngr would love! i made this very thing for a hackathon once ;p at the time models struggled hard with framer syntax but i bet all of that's solved™ with modern opus etc.

@nicoalbanese10 very cool!

3 months ago I started building a coding agent that runs in the cloud. It's since written every line of code I've shipped, including itself. Today, I'm open sourcing it. Introducing Open Agents.

@ridd_design notion portfolio is fine imo — i'd much rather designers spend their time shipping real products than maintain a fancy personal website (although bonus points if you can do both ofc)

if you're applying to roles right now with a Behance/Notion portfolio you're not doing yourself any favors 😬 market is too competitive to not have a website that is a personal expression of who you are as a designer

@jacobgold @MerrillLutsky @graphite @cursor_ai sweet, thank you! if at any point you need cloud x graphite testers... lmk (come hang in the #mainframe-cursor channel!) we're heavy users of both so looking forward to it!

@vinvan @MerrillLutsky @graphite @cursor_ai we're kind of figuring this out right now. short answer is that mostly we either don't, or do what you described. actively working to enable this more first-class

.@MerrillLutsky @graphite any tips for making agents use graphite? they seem to struggle with the cli and fall back to git, override work, or rebase manually. i haven't heard much about the mcp since last october. is that still the move? an official skill would be cool!

@jacobgold @MerrillLutsky @graphite also curious: how do you use stacks with @cursor_ai cloud agents? right now i tell every cloud agent "hey, i have a stack with the following branches: [....], check them all out and run gt track --force from the tip" that works but wonder if there's an easier way.

@jacobgold @MerrillLutsky @graphite thank you! is there anything you put in agents.md as well to make it gravitate towards these skills?

@jacobgold @MerrillLutsky @graphite hell yeah. tysm!!

@vinvan @MerrillLutsky @graphite github.com/goldjacobe/gra…