Y Combinator
Make something people want.
Start new thread
Nika

Y Combinator offers 7 startups ideas they want to fund (Spring 2026)

by
minimalist phone: creating folders
•

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 Train

Which 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? 👀

4.7K views

Add a comment

Replies

Best
Princewill Okafor

Oh no wonder i got rejected from YCombinator😩 ...... My product isn't in this list.

My Vote for the list would be [2], [5], [4], [6]

I would add AI to create assessment from Learning materials & video

I dont understand the other product category

Nika

@oprincew What was your product you tried to apply with?

Praney Behl

I am on to AI-Native Agency, already working on it. Will share updates soon.

Nika

@praney_behl Are you applying to YC?

Praney Behl

@busmark_w_nika Not planed yet, I have been bootstrapping so far.

Praney Behl

@busmark_w_nika Always looking for meaningful partnerships to help scale GTM though

Nika

@praney_behl Are we already connected on LinkedIn? :) for sure

Abdul Rehman

I’m really interested in AI Guidance for Physical Work. Most AI startups target digital workers, but helping people in real-world jobs could create massive real-world value.

Nika

@abod_rehman I am mostly curious about that one. Do not know why, but I could see there some collab with humanoids and AI.

Miranda Fair

Definitely 1. We've been building an AI-native tool to help teams implement Scrum and ship faster, and we realized there’s another gap upstream: generating the right ideas in the first place. We're experimenting with something that helps teams rapidly ideate, prioritize, and specify changes based on real data (very much in the "Cursor for PMs" spirit). Once you know what to build, it flows nicely into Scrum execution. We've been testing it with fun prompts like “what new roller coaster should Disney World build” and the outputs are surprisingly cool!

Nika

@miranda_fair1 Wait! Do you already have anything in production you want to use for applying to YC?

Dario Sansano

We are currently working on solving point Nº 10 at @NeuroBlock. We created a no-code AI Lab so that anyone can train their own small, specialized AI models using their proprietary data, without needing any technical skills.

We abstract everything, from GPU configuration to data processing, pipelines, and hyperparameter tuning. Users simply upload their raw datasets, and we handle the entire process end-to-end. They go from raw data to their own small, specialized AI model, which they can run in our cloud, download and run locally, or deploy in their own dedicated cloud instances.

We’ve just applied to YC. We truly believe this is the direction AI training needs to take, and we have already launched here and secured our first paying users.

This is an incredibly exciting field for us. We strongly believe that foundational models, as we know them today, will shift toward in-house models developed within companies and organizations etc, built on top of proprietary data and internal know-how, rather than relying solely on external massive, generalized models and the pursuit of so-called AGI that everyone accesses through a "single" API.

Nika

@dario_sansano wow, wishing the best! :) Let us know whether they approved it! :) It could be a cool story :)

Aayush

Nika! Thanks for the post. We have been bullish on AI-Native Agencies since a while! RFS was such pleasant timing, especially since the last one was in Fall 25 (and we got a No from the interviews for a different idea. Very valid feedback which dictated our pivot). I have led Content and Growth for a YCW20 before, handling mid-tier content creators, and realized how burnt out creators are in admin work around brand deals (outreach, discovery, invoicing etc.). Alas, it's their biggest revenue source and agencies charge them 25%+ . Building an AI talent management agency to solve for this. Making users money already! https://usesnippet.app

Nika

@aaupadhy Will you also offer a freelance package too? I think that this is not only a problem of creative agencies, but freelancers who are also short of time and busy by doing real creative work – they do not want to be halted by administrative tasks.

Nikhil Shahane

We're in the process of reinventing our agency to be AI Native. It's the future of Agencies of all kinds and it's the perfect combination of being 'Techno-Human'. The humans will bring the 'taste' and the AI will do the heavy lifting.

As I've mentioned, the ultimate barrier to any progress/evolution is the rate at which people adopt anything. I work primarily in advertising and marketing - and aside from the buzz around the term AI - most people have NO idea what it actually entails or the rate of progress being made. This is why the AI-Native agency is most likely going to be 'new' and not an incumbent. (At least the majority of the incumbents aren't going to make it.)

Nika

@nikhilshahane Just out of curiosity, which topics related to AI and digital agencies are interesting to you? :) I would like to contribute more in these terms.

Nikhil Shahane

@busmark_w_nika The way I see it is that agencies have 1 job - to add value to their clients in the field that they are operating in. Eg: For brands & businesses, it's creative value in the form of advertising or marketing strategies and tactics to increase revenue (and there will be proxies for this - your general brand metrics fall into this category). Or in another category - business strategic consulting agencies will need to help with ops / GTMs, etc... and each category will have its own thing (PR, Talent Management, Operations, etc... etc...)

And in any agency there are 2 broad work streams - the thinking (taste) and the execution (ops). In most businesses software has already helped optimize the ops bit. But most agencies haven't made that shift yet. That's one clear part where I see the AI-Native agency coming into being.

My prediction is that there will be many more small boutique agencies that bring immense 'taste' to the table that will no longer be 'small' because AI (software) will give them the ability to scale.

Or I could be wrong and we'll not need agencies at all. :D

nim

id add "ai-native content consumption" to the list tbh. the way people read (or dont read) the things they save is broken. i built speakeasy which turns article urls into audio — bc most people have way more to read than time to sit and read it. audio unlocks commutes, gym, cooking etc

feels underinvested relative to how much content overload is a real problem. podcast apps are huge but "listen to any article" hasnt had a proper ai-native moment yet

Nika

@sup_nim This is a problem of spanned attention. Peope should try a stronger will and reading books imo :D

Alexey Glukharev

The 7 looks like interesting and doable. I think we at OceanMindApp made one step in this direction but defensively need another startup not a pivot for it.

Good call, let’s research!

Nika

@alexeyglukharev are you gonna launch something? 👀

Michael Vavilov

Well, Anthropic filled this #1 gap with Claude CoWork imo

First
Previous
123