Y Combinator offers 7 startups ideas they want to fund (Spring 2026)
As usual, Y Combinator came up with segments that are worth investing:
1. Cursor for Product Managers
2. AI-Native Hedge Funds
3. AI-Native Agencies
4. Stablecoin Financial Services
5. AI for Government
6. Modern Metal Mills
7. AI Guidance for Physical Work
8. Large Spatial Models
9. Infra for Government Fraud Hunters
10. Make LLMs Easy to TrainWhich of the following do you think are the most interesting?
OR
Would you add to the list, maybe something that is more worth funding?
Also, are you going to apply? đ
4.7K views
Replies
minimalist phone: creating folders
@siarhei_ai No. 7 :)
ProblemHunt
Hi Nika! As soon as I saw the first request on the list, I couldnât pay attention to the rest anymore. It really feels like «Cursor for Product Managers» is something thatâs very needed and could take off strongly. But letâs see what time shows đ
Weâve submitted the ProblemHunt application for the second time.
minimalist phone: creating folders
@gostroverhov uuu, i have my fingers crossed for you! :)
ProblemHunt
@busmark_w_nika đ€
Scade.pro
@gostroverhov good luck, Boris. Nice product name :)
ProblemHunt
@nastassia_k thank you, Nastassia :)
Great breakdown! Iâd probably vote for #1 (Cursor for PMs) as the most immediate need, but #7 feels like the biggest long-term shift for the workforce.
Iâd add Personalized AI Tutors to the listânot just for coding, but for general skill acquisition. We need better ways to keep up with all this tech!
Good luck! đ
minimalist phone: creating folders
@jakub_malek1Â I think that 7 is the most revolutionary and imagine including robots. That would be pretty awesome (and scary) :D
@busmark_w_nika Haha, exactly! Itâs a mix of excitement and 'Terminator' vibes. :D
Speaking of the AI Tutorâwe actually applied to YC with it recently. We didn't make the cut this time (super competitive batch and slightly different market focus), but weâre not letting that stop us. Weâre iterating on the product and definitely applying again! đ„
minimalist phone: creating folders
@jakub_malek1Â Wishing you good luck. To be honest, I do not know their criteria, but many people post on X about how they were rejected (it is true that mostly they were solo founders). Hopefully, your product will be appealing to them! :)
@busmark_w_nika Yeah, their exact criteria aren't public, but from what I've heard, the acceptance rate can be as low as 1 in 100. So the odds are tough, but we're definitely not giving up! đȘ
Iâm putting all my bets on AI for Government! The dream of AI finally killing off soul-crushing bureaucracy and processes stuck in the 90s is just too good to ignoređ„č
My only worry is that most governments are still have what to do with basic digitalization, let alone AI :D
But yes, itâs definitely the biggest area for a massive upgrade!
minimalist phone: creating folders
@valeriia_kuna I think that this is more likely for China. Europe? eeeh :D Not so sure.
@valeriia_kuna It's a big bet, but goes against the psychology of how the government operates. Most governments are about control. AI will reduce that. Same reason why blockchain/cryptography based tech has always been blocked by governments.
Having said that, I'm with you - I hope for a day where we don't have to deal with soul-crushing bureaucracy and red-tape.
minimalist phone: creating folders
@fitnessrefined People in bureaucracy will not be happy, that's for sure.
AI-native agencies stand out to me. Itâs basically: pick a boring but valuable niche, map the workflow, then replace chunks of it with agents that run autonomously. The hard part isnât the tech, itâs the scoping.
minimalist phone: creating folders
@solomon_nwabuoku To be honest, I am not so technical, but for me, the execution would be tricky :D
@solomon_nwabuoku Running autonomously is hard. Even claude cowork dont support agent finding their next tasks to do on their own.
@solomon_nwabuoku I'd wager the hard part will be getting the 'Taste' bit right. That's what agencies are supposed to bring to the table. And even before this AI wave - most agencies had lost that.
vibecoder.date
Mills as in machining equipment?
I think stablecoin based financial services that focus on being tax compliant would let local businesses take crypto without the regulatory nightmare.
minimalist phone: creating folders
@build_with_aj Now, I am pretty hesitant about crypto. I invested, and it all went down. đ It happens to me every time. LOL.
interesting list but what's missing is #11. Tools to figure out what's actually worth building in the first place.
everyone's racing to build AI products but nobody's asking what problems people actually have. i might be crazy but i went the opposite direction. I built something with no AI at all.
frikt is an app where people post their daily frustrations in one sentence. others say same. patterns emerge. the idea is that before you build anything, you should know if enough people actually have that problem.
with AI making it so easy to build, the bottleneck isn't building anymore. it's knowing what to build. that's the gap i'm trying to fill.
minimalist phone: creating folders
@karolisbuilds I can say that there is a page that captures real problems so people can build a product according to that â it is a @ProblemHunt
Stellify
I'm building an AI code editor (stellisoft.com). From what I've read it doesn't fit directly into those categories but I still intend to apply. Are you applying Nika?
minimalist phone: creating folders
@stellify_software Me not :D I do not have that ambition (and no idea what to build) :D Do you have co-founders? I have heard that co-founders are also a factor that influences the decision to approve
Hi Nika!
I think AI for physical work is really interesting, though high cost of entry, far horizon and a lot of vaporware right now.
Even more fascinating is infrastructure tech enabled by AI, in part for the science and engineering boost, and in part on the massive capital generated by AI adoption. Stuff like energy, orbital datacenters, semiconductors, drug discovery... But again capital intensive and a bit far in the horizon.
AI agencies although less monumental also seem very interesting to me, the concept of being able to fully delegate not just tasks but full plans and proactive execution... damn, that's the kinda sci fi I wanna see.
And yes! I am applying with Rkive AI, which is pretty much an AI content creation agency.
minimalist phone: creating folders
@alberto_luengo Do you apply as a co-founder or solo?
To be honest, I'm not sure what to expect from AI agencies. :D
@busmark_w_nika Solo! I have two employees but no co-founders.
Tbh I also don't know what to expect from AI agencies, I think we will see a lot of plain agencies with an AI chatbot sprinkled on top đ I would like it if these "agencies" had all the specialist agents you could need, working for your outcomes proactively, with all your context, etc. But I think it's a bit too vague and conceptual to be useful, and I don't like the shift towards zero visual UI and zero manual control because it seems impractical. Lately I brand Rkive as the "Content Creation OS" for that reason. Really ai-native but also with an entire UI and manual control built around it.
I suspect AI is gonna move in that direction. Sticking to the current "agencie" experience seems like an accounting firm back in the day adopting excel to replace manual accounting, but still offering their reports via handwritten letters.