Cursor for Product Managers Andrew Miklas AI is great at writing code, but product success depends on deciding what to build. We think there’s an opportunity for an AI-native system focused on helping teams figure out what to build, not just how to build it.

The way startups are built has shifted quickly. We're excited about a range of startup ideas for AI-native companies that can now be built faster, cheaper, and with more ambition than ever.

AI-Native Hedge Funds @charlieholtz In the 1980s, quantitative trading looked fringe; today it’s standard. We’re at a similar inflection point with AI, and the next great hedge funds will be AI-native, using agents to analyze filings, calls, and markets at machine scale.

AI-Native Agencies @aaron_epstein Agencies have always been hard to scale. AI flips this model by letting firms use software internally to deliver finished work at higher margins, turning agencies into software-like businesses that can scale far beyond today’s service firms.

Stablecoin Financial Services @DaivikGoel Stablecoins are rapidly becoming critical infrastructure for global finance, yet much of the financial services layer remains unbuilt. This creates room for services that offer the benefits of DeFi. Now is the perfect time to build something in this space.

AI for Government @t_blom AI has made submitting forms dramatically easier, but governments still process many of them by hand. There’s a major opportunity for AI-native tools that modernize government operations—hard to sell, but powerful once adopted.

Modern Metal Mills @zanehengsperger Reindustrializing the U.S. isn’t just about labor or geopolitics—metal mills are slow because they run on outdated systems. Modern software creates an opportunity for mills to cut lead times and costs while improving flexibility and margins.

AI Guidance for Physical Work @dflieb Much of the AI conversation focuses on which desk jobs will get replaced. But for physical work, AI can't yet act in the world. With multimodal models and ubiquitous hardware, there’s an opportunity to give workers live AI coaching that dramatically reduces training time and unlocks skilled labor at scale.

If you’re interested in working on any of these ideas, you should apply to YC. The deadline for applying to the spring batch is February 9th at 8pm PT:

But wait, there's more! Large Spatial Models - Ryan McLinko Recent AI progress has been driven by LLMs, but true breakthroughs will require robust spatial reasoning. There’s an opportunity to build foundation models that treat geometry and physical structure as first-class primitives, and enable AI to reason about the physical world.

Make LLMs Easy to Train - Gabriel Birnbaum Training large language models is still surprisingly difficult. As post-training and model specialization become more important for a wide array of businesses, we'd like to see products that make LLM training easy.

Infra for Government Fraud Hunters @garrytan We want to fund startups that bring government fraud investigation into the modern era. There’s a major opportunity today for systems that turn whistleblower tips into complaint-ready cases, dramatically speeding fraud recovery for law firms and government agencies.

@amiklas Hey @ycombinator & @amiklas… we heard you’re looking for us 👀 We’d be happy to show you what we’ve built so far, and we may even have a little room left in the round. 😉 DM and check out

@amiklas @from021_ !!!

@amiklas Traycer captures dev intent better than anyone (asks questions if need be) builds a detailed spec breaks it down into tickets Orchestrates execution (Bart Simpson for the win) Verifies changes and ensures that the code is actually aligned with what you are building.

@amiklas It’s already here though: @modemdev

At @uselayr this is exactly what we are solving: teams are not short on data, they are short on decision clarity. Product context is scattered across Slack, Jira, docs, and feedback, so prioritization becomes noisy. We help teams turn that signal into evidence-backed decisions on what to build next.

@amiklas @amiklas 🤚🏾 This is something @usereavil built for product teams. We also use customers feedback as part of our data source for deciding what to fix, features to build and level of impact is has.

@amiklas We built in 24 hours for a hackathon in Hamburg last weekend. Its vision is literally this + a lot more goodies.

@amiklas Then yall should def check out @tryDotted - literally cursor for product managers ⚡️@DunniAbiodun has been building a madness with this one 🤖

@amiklas @shreyas oh no the tarpit of a tool for PMs

@amiklas

@amiklas @arielmathov

@amiklas oh shit @lukalotl there you go

@amiklas Isnt claude cowork the cursor for product management? Atleast thats how ive been using it

@amiklas I solved this. Free tool here:

@amiklas 100% Input/context is key now, with current output velocity. we are building in this space we would love to hear from you 🫡

@amiklas

@amiklas Upload interviews + usage notes -> ranked ‘what to build next’ -> outputs PRD + ticket breakdown for coding agents. Demo:

@amiklas @usehamster

@amiklas What does a Cursor for PMs actually mean? For us: • Understand messy product context • Extract decisions and action items • Update tickets automatically • Reduce manual coordination PMs should guide the work, not constantly reconcile it.

@amiklas @async

@amiklas But Windsurf has a Plan feature, this is not needed. It's just a feature on AI IDEs already. Plus Linear is working fine!

@amiklas 👋

@amiklas Yeah we are currently building this product (cursor for product managers) for an fintech startup.

@amiklas We built this and we think its pretty great. Used it to help YC companies figure out what to build next:

@amiklas We built this ! Check out and @HelmPM_

@amiklas I think this is what @linear is working towards.

@amiklas I’ve been pondering how to collapse customer service into product insights to help product managers get unfiltered feedback from users

@amiklas I totally agree! We have to solve problems that can actually make a difference

@amiklas Well if that's the case you should checkout :

@amiklas im building it and taking an approach with mix of startups teams, experts and customers with ai agents for product discovery

@amiklas This is something I have been building called Decision X. It doesn’t plug user interviews yet, but it already helps on scenario comparison. “What to build” is the next step. Funny thing YC is into that.

@amiklas almost done building it

@amiklas @dharmeshba