Every few months, Y Combinator does something no other investor does: it publishes exactly what it wants to fund.
No vague thesis. No “we back bold founders.” A list. Names of problems. Names of industries. Written by the partners themselves.
They call it the Requests for Startups. And the Summer 2026 edition just dropped — 15 categories, and one opening line that tells you everything:
“AI has stopped being a feature and started being the foundation.”
This isn’t a wishlist. It’s a signal. And if you’re building in Pakistan right now, some of it is pointing directly at you.
What YC Actually Is
Before we get into the list: if you’ve heard the name Y Combinator but aren’t totally sure what it is, here’s the short version.
YC is a startup accelerator based in San Francisco. Since 2005, it has funded over 5,000 companies — including Airbnb, Stripe, Dropbox, DoorDash, and Reddit. Every company it accepts gets $500,000 in exchange for 7% equity, plus three months of intensive mentorship, office hours with partners, and access to one of the most powerful alumni networks in the world. The whole thing ends with Demo Day, where you pitch in front of hundreds of investors.
The acceptance rate? Under 1%. More selective than Harvard.
And yes — it’s open to founders from anywhere in the world. Pakistani founders have been in it before. More on that below.
The Summer 2026 List — All 15 Categories
Here’s every category YC published for its Summer 2026 RFS, straight from ycombinator.com/rfs:
- AI for Low-Pesticide Agriculture
- AI-Native Discovery Engines
- AI-Native Service Companies
- AI Personalized Medicine
- Company Brain
- Counter-Swarm Defense
- Dynamic Software Interfaces
- Inference Chips for Agent Workflows
- Lunar / Industrial Moon Capabilities
- SaaS Challengers
- Semiconductor Supply Chain Software
- Software for Agents
- Space Electronics
- Stablecoins / New Financial Infrastructure
- Startups That Want to Sell to Huge Companies
Eight of these are deep hardware and capital plays — space, defense, chips, semiconductors. Hard to build from Lahore, at least at this stage.
The other seven? Pure software, or services. Small teams. Global markets. Equal footing.
The Pakistan Problem
Here’s the thing nobody says out loud: YC doesn’t just fund ideas. It funds founder-market fit. The founders who built Stripe were frustrated by how hard payments were. The founders of Airbnb couldn’t afford rent. The insight came from living the problem.
Pakistan has problems. Real ones. Documented ones. The kind YC is explicitly asking for startups to solve right now.
Let’s go category by category.
1. AI for Low-Pesticide Agriculture
YC’s pitch for this category reads like it was written about Pakistan’s farming sector.
The case: pesticide overuse is destroying yields, contaminating soil, and trapping farmers in a cost spiral they can’t escape. Farmers spray more, get diminishing returns, costs go up, margins collapse. YC wants startups using AI and robotics to break this loop.
Pakistan is the 8th largest agricultural country in the world by cultivated land area. Agriculture makes up roughly 24% of our GDP and employs over 37% of the labour force. Pesticide misuse here isn’t hypothetical — it’s a documented crisis, affecting cotton, wheat, and rice at scale.
A founder who grew up watching this. Who worked in agriculture extension, or agri-fintech, or supply chain. Who understands the actual farmer — not the one in a YC slide deck — has a kind of insight no Stanford graduate can fake.
This is a YC-sized market. And Pakistan has YC-sized founder-market fit for it.
2. AI-Native Service Companies
This one is quiet. Don’t sleep on it.
YC’s thesis here, written by partner Gustaf Alströmer: the next wave isn’t companies that sell software tools — it’s companies that do the work entirely. AI-native firms that replace services, not just improve them. Think AI-powered accounting firms, compliance shops, insurance brokers, healthcare admin. You don’t buy the software. You buy the outcome.
Pakistan exports services. That is literally what our tech economy runs on. BPO, accounting, compliance, legal work, software development — we’ve been doing this for decades. The talent infrastructure exists. The English proficiency exists. The price advantage exists.
The founders best positioned to build AI-native service companies replacing global services are the ones who already know what those services look like from the inside. That’s us.
3. Company Brain
Tom Blomfield’s (founder of Monzo) category. The idea: every company has critical knowledge scattered across email threads, Slack, support tickets, and people’s heads. AI can’t automate companies until that knowledge is structured and accessible. The startup that builds this — a living, updatable “brain” for how a business operates — is, in his words, something every company in the world is going to need.
This is pure software. Two founders. Laptop. Global market.
Pakistani developers are among the most active contributors to the open-source tooling that sits under exactly this kind of product. This category has no geographic moat. The playing field is flat.
4. SaaS Challengers
YC wants startups that go after legacy SaaS incumbents — Salesforce, SAP, Oracle, the giants — using AI to deliver enterprise-grade software at a fraction of the cost. The insight: AI-native teams can now ship what previously took 200-person engineering orgs to build.
Pakistani software companies have been undercutting global SaaS pricing for years. We know how to build fast and lean. The difference now is that “fast and lean” can mean better, not just cheaper.
5. Startups That Want to Sell to Huge Companies
The most underrated category in the entire list.
YC is explicitly saying: Fortune 100 enterprise sales is now accessible to year-one teams. AI lets two founders ship enterprise-grade products. The barrier that used to require massive engineering teams and decade-long sales cycles is collapsing.
Pakistan has produced thousands of software engineers who have worked inside global enterprises — as contractors, outsourced developers, consultants. They know the workflows. They know the pain. They know the buyer.
That’s not a credential. That’s founder-market fit.
6. Stablecoins / New Financial Infrastructure
$31 billion. That’s how much Pakistan received in remittances in 2024. The highest in the country’s history. Every dollar of that moved through a system that is slow, expensive, and riddled with middlemen.
Stablecoins and new financial infrastructure are YC’s bet on rebuilding the plumbing of global money movement. Pakistan is one of the top remittance-receiving countries in the world. The problem is not abstract here — it’s felt in every household waiting for money from Dubai or London or Riyadh to clear.
A Pakistani founder building stablecoin infrastructure for remittance corridors isn’t chasing a trend. They’re solving their own family’s problem. YC loves that story.
It’s Been Done Before
If you’re thinking “this isn’t for people like us” — Pakistani founders have been in YC since 2014.
Muneeb Ali (LUMS → Princeton PhD → YC S14) co-founded Blockstack, now Stacks — a smart contract platform anchored to Bitcoin that went on to complete the first SEC-qualified crypto token offering in US history.
Waqas Ali and Sidra Qasim took their artisanal shoe startup Markhor into YC S15 — the first Pakistani company based in Pakistan to be accepted. They came from rural Pakistan, made it onto a Kickstarter that reached 508 backers in 32 countries, applied to YC twice (first rejection in 2012, accepted in 2015), and went on to build Atoms — a sneaker company backed by Kleiner Perkins.
Umer Adnan took Cowlar — a smart collar for cows built in Pakistan — into YC W17. Made it to the top 7 at Demo Day. In 2017. Building AgTech. From Islamabad.
In Winter 2022, four Pakistani startups got in at once: Elphinstone, Mahaana Wealth, Rider, and Markaz. All building for Pakistan. All accepted.
The door has been open. It’s still open.
What Comes Next
This is the first piece in our Pakistan to YC series — a step-by-step guide for Pakistani founders who want to apply.
Over the next few weeks we’re covering:
- What YC actually looks for (and what it doesn’t)
- The Pakistan opportunity map — which categories have the highest founder-market fit
- The application breakdown — what each section actually means
- The logistics reality — visa, incorporation, what moving to SF for 3 months actually looks like
- Stories from Pakistani founders who’ve been through it
The next YC batch is Fall 2026. The on-time deadline is July 27, 2026. Applications are open now at ycombinator.com/apply.
You have time. Use it.
