Launching today
folk
the AI in your texts that gets stuff done
326 followers
the AI in your texts that gets stuff done
326 followers
Folk is your AI friend that lives in any messaging app: iMessage, Telegram, Discord, and more. It joins meetings, plans your date, knows where you are, and gets smarter with every conversation. The longer you use it, the more it knows you. And now it's multiplayer, so you and your friends can team up and reach your goals together.
Yoo Product Hunt 👋
I'm Arlan. I dropped out of high school, did research at Stanford at 16, built multiple apps, went through Y Combinator, and have raised over $6M.
I spent the last year going deep on context and memory. I tried everything out there, and two things kept bothering me. First, none of them actually grow with you. They're smart in the moment, but tomorrow they don't remember yesterday. You're always re-explaining yourself, always starting over.
Second, none of them are actually proactive. You're always the one going to them, they never come to you. A lot of tools claim to be proactive, but really it's just a cron job firing on a schedule, and a timer isn't proactivity. Real proactivity is showing up at the right moment on its own, texting you exactly when something matters, without you setting anything up.
Folk is different. It lives inside the messaging apps you already use and it's built from the ground up to integrate into your life, not just assist when you ask.
Here's what that looks like in practice:
It knows where you are. Folk tracks your location and acts on it. Walk past a grocery store with items on your list? It'll ping you. It's not a map, it's context.
It joins your meetings. Folk's notetaker drops into every Google Meet, transcribes everything, and builds on what it learns so it always knows your world.
It remembers everything and gets smarter over time. Folk builds a growing memory of your life. The longer you use it, the more it knows you. That's the whole point.
It's multiplayer. This is the part I'm most excited about. Add friends who are already on Folk, and your folk and theirs can do things for each other, like book a table, share a doc, or check a calendar. One way, and only ever with a tap to approve.
And it does real things like research, coding help, date planning, booking restaurants, and tracking flights, all from the chat thread you're already in.
We're giving Product Hunt users 15% off today with code FOLK.
Two questions I'd genuinely love your take on:
What part of your life do you most wish an AI could just handle for you?
What would make you actually trust an AI with something personal?
Thanks for being here 🙏 Arlan
folk
@arlanrakh @tehreem_fatima5 thank you! one of the most thoughtful answers here
the 'gets smarter with every conversation' claim is the one I'd want to pressure test. most AI assistants that claim this either mean they store a growing context file or they fine-tune on your data, and those are very different things with very different privacy and quality implications. what's actually happening under the hood when Folk learns from you and where does that data live
folk
@ansari_adin good pressure test. folk is neither of your assumptions. no fine-tuning and no dumb growing context file. its a structured knowledge graph running in your own isolated instance. think nodes and edges back when you took DS&A in college. the facts are nodes, and the edges form over time as the agent connects what it learns about you. super simple. u building in the space?
@ansari_adin fair question. neither of those.
we don't fine-tune on your data. what actually happens: after conversations, folk extracts durable facts about you (preferences, people, context) into a private memory store tied to your account. at the start of each session it pulls the relevant pieces back in, so it doesn't need your full message history in the prompt every time.
Prava
Amazing, can I install custom plugins/skills in folk? How does the integrations work?
I was looking to connect my card with it for lunch orders
folk
@kaiserrr folk will NEVER store your card. ima keep it a buck, theres no "connect your card" in folk today.
INSTEAD, Folk drives the order in its own browser, and when it hits the pay step it pings you (human in the loop amirite). Folk can stay signed in to the ordering site after the first time, so you can legit send a msg such as "folk, order my usual"
is a saved-payment option something youd actually want? this is the kind of thing we'd only ship if we could keep it as safe as everything else in folk.
@kaiserrr yes u can! just ask folk to connect you to something + u can create custom skills as well :)
@kaiserrr You can just ask your folk to install skills or integration for you. Also, for integrations and MCP servers, you can set them up manually on dashboard if you want
mcp-use
yoooo! congrats on the launch @arlanrakh
do you support custom integrations / MCPs?
@pederzh yessir, just tell it to connect to any type of mcp or go to https://www.getfolk.app/dashboard -> connections -> custom if MCP requires OAuth
Product Hunt
I'm pretty into the iMessage -> Assistant workflow and am currently using @Poke.com pretty heavily. How does Folk compare and differentiate from other assistants like Poke?
@gabe hey, love that. poke is genuinely good, they nailed the proactive-texting feel.
geo. folk actually knows where you are and acts on it. walk past the grocery store with stuff still on your list and it pings you. change cities and it adjusts asw (tested it when going to nyc lol)
multiplayer. add friends who are on folk, and your folk can work with their folk. "ask maya's folk to find a sushi spot friday" and the two agents coordinate it, one-way, and only ever with her tap to approve. your assistant stops being a silo.
we also have bunch of other things like meeting note taker, custom memory system that supports 3 memory types, and slightly different agent architecture!
Product Hunt
@arlanrakh oh very interesting! Going to give it a try
@gabe arlan@nozomio.com for any feedback (will give u my personal cell asw if u are ok w that just shoot at an email)
thank u so much!
The memory part is what I want to understand better, how does it actually build context over time? is it pulling from message history or does it need explicit inputs from us?
folk
@munis_abbas automatically! just by talking to it. you talk, folk listens, and keeps the things worth keeping (who matters, what you like, the plans youre nursing) as nodes, then connects them over time, with a quiet overnight pass where it thinks about you.
you can be explicit if you like ("remember my sister is vegetarian"), but mostly it just gets to know you over time. i wouldnt force it. mention one oddly specific thing today and itll resurface a week later..
the exact moment i fell in love with Folk was when it was the second one to tell me happy birthday!
the first one was obviously the team at Nozomio haha.
@munis_abbas both, but not in the way you'd expect from raw chat logs.
folk keeps a curated memory of you (preferences, people, habits, open loops) that gets updated as you talk. it also has access to recent conversation history and can search past sessions when something relevant comes up. you don't have to manually teach it everything, but you can. if you tell it something explicitly ("remember i hate cilantro"), it sticks. over time the useful stuff compounds and the noise gets dropped.
Really enjoying the uptick in services that live in iMessage. One thing I've noticed with all of them - folk included, I'm afraid - is that latency varies wildly... like, sometimes it'll start responding as soon as I send a message, other times - like right now, as I stare at my phone - I sent a message minutes ago and it's not even typing, but it DOES show "read".
Is this endemic to iMessage's infra? what's up with that?
also, @arlanrakh - what's folk powered by? you down to give us a peek under the hood?
congrats on a highly successful launch, thx for the hunt, @garrytan !!
folk
@arlanrakh @garrytan @grey_seymour thank you for the sharp comment!
the read fires instantly because thats handled the second your message lands. the silence after is on us unfortunately, not apple, behind that read your folk is sometimes waking its cloud computer, pulling memory, or already opening the browser. we were thinking of dropping in little "on it, pulling your data" notes, but they clutter the imessage thread.
under the hood, everyone gets their own 24/7 cloud computer with a real browser and a knowledge-graph memory, and we route across frontier models.
if you have any questions definitely let me know
@grey_seymour hey grey, fair call on the latency and you're not imagining it.
the read receipt and the reply are two different pipes. "read" is imessage acknowledging delivery. the actual response has to wake your folk's computer, load your session + memory, run the model, then send back. if the sandbox was cold (nobody texted you in a while), that wake-up can take 2s before you even see typing. when it's warm, it's basically instant. w
and thanks for the kind words on the launch, means a lot.