co-founder @mainframe. prev: @figma @diagram @facebook.
NYC
Joined August 2013
- Tweets4.6K
- Following1.8K
- Followers9.8K
- Likes32.1K
the future of design isn't drawing pictures and developer handoff as a step shouldn't exist.
Vincent van der Meulen retweeted
we're starting to test a new app with designers @mainframe
DM or reply if you want to try
i've been shipping prs from my runs, showers, and commutes by calling @bot.
get yourself an always available eng manager. go into debt if you have to – at New York, NY
how to use @bot for engineering and get a ton done:
1. create a (cursor) project for every big workstream. i have ten, such as a “mcp project” and a “bugs project”
2. make sure every project spins up cloud agents with /poteto-mode
3. in grok bot, make an eng bot whose sole job is routing requests to the right project.
4. now voice call (!) your primary bot and braindump all your tasks.
your tasks will automatically end up at the correct projects, and result in sweet prs :)
you get all the benefits of projects + pstack, but don’t have to think about anything!
how to use @bot for engineering and get a ton done:
1. create a (cursor) project for every big workstream. i have ten, such as a “mcp project” and a “bugs project”
2. make sure every project spins up cloud agents with /poteto-mode
3. in grok bot, make an eng bot whose sole job is routing requests to the right project.
4. now voice call (!) your primary bot and braindump all your tasks.
your tasks will automatically end up at the correct projects, and result in sweet prs :)
you get all the benefits of projects + pstack, but don’t have to think about anything!
one underrated hack is to let your @bot manage your screen time.
e.g. my grok bot has access to my rescuetime so it can change what websites are blocked, checks in daily about # hours worked, etc.
the paid partnership label makes any brand look so cringe that i’m not sure the eyeballs are worth it
why shipping 100s of prs makes sense
there's always something to improve in your product's codebase. if your goal is to make customers happy as fast as you can, it makes sense to fix those things the moment you find out about them. we're no longer limited by intelligence!
some of this is of course the result of the new way of working. ai is a net-benefit for codebases, but leaves behind more residue as people are less in tune with low risk areas. which you then have to squash by shipping more.
of course it's a little bit different for frontend/design, where the (human) thought required for the average fix is higher. but in backend land, there are always obvious bugs waiting for you to eliminate them.
i think many people would benefit from focusing on speed! it's the equivalent of the running advice to "just run more."
unless you cheat your way out, it's hard not to get to a better product/codebase/engineering skills by simply trying to put up more prs.
we're in the brief period where models still need our slight handholding and energy, which is why this advice still feels uncomfortable. but soon enough models will make the rate at which we used to evolve products feel dull.
merged 145 prs so far today! would recommend the @poteto playbook:
1. get a lot of tokens
2. create a project (orchestrator) for every workstream. only talk to them, via ui or grok bot
3. tell your projects to spawn cloud agents in poteto mode
4. verify with cross model reviews
i think token scarcity is secretly causing people to underperform.
you can't do your best work if you're worrying about limits, picking dumber models than you actually need, and holding back tasks until resets. – at Manhattan, NY
i'm noticing it in myself and my decision making.
it reminds me a lot of this book (banger btw): amazon.com/dp/0805092641 – at Manhattan, NY
it should be your #1 priority to make yourself feel token abundant.
i myself finally slopped together a setup that gives me unlimited tokens again.
unless your employer likes paying tens of thousands of $, this is annoying, recurring work that takes time. but worth it! – at Manhattan, NY
@poteto how do you decide when to create a project?
i read that you have ~10 parallel projects and was curious why you don't have a single, big one!
since the benefit of projects seems to me that you don't have to worry about routing. – at Manhattan, NY
e.g. i imagine for perf work one could make their persistent project spin up a smart agent (e.g. opus 5.5 xhigh) and route all perf work there. instead of create a 2nd project?
but not sure what tradeoffs are/what cursor does internally!
also cc @alexirobbins! – at Manhattan, NY
the labs need facebook's '2g tuesdays'
but instead of slow internet, one day a week employees get normie token limits – at San Francisco, CA
👀 we have time to visit a few more companies today!
dm if you're a small sf startup and want to give your thoughts on an early demo!
we'll come to you :) – at San Francisco, CA
i saw the future and now the present is indistinguishable from the past
@mainframe boys did it again 👏
Vincent van der Meulen retweeted
if you’re a designer at DevDay @vinvan and I would love to meet you!