Y Combinator, Techstars and Antler Applications: Requirements, Acceptance Rates and How to Answer the Questions

Y Combinator application requirements are the most searched topic in startup fundraising for a reason: the programme has backed more than 5,000 companies, funds 140–200 per batch, and its form has become the template that Techstars, Antler, Entrepreneur First and 500 Global adapt. This guide, part of the Kurums Venture Capital hub, sets out how each programme’s application works as of October 2026 — batches, deal terms, acceptance rates, form questions, the video and the interview — compares them in one table, and explains how to answer the questions that actually decide outcomes.

How does Y Combinator’s application process work in 2026?

YC now runs four batches a year — Winter, Spring, Summer and Fall — each about three months long and in person in San Francisco. You apply online, submit a one-minute founder video, and if shortlisted take a ten-minute video interview with two or three partners.

The YC application page currently shows the Winter 2027 batch open, with an on-time deadline of 2 November 2026 at 8pm Pacific, decisions by 11 December for on-time applicants, and interviews in November and December. The batch runs January to March 2027 and starts with a three-day in-person kick-off. Late applications are still read but without a promised response date, and third-party guides consistently report that late applicants have lower odds. YC also offers early decision, which lets founders apply now for a later batch if they are not ready to move in January.

The move to four batches, made in 2025, changed the maths. Recent cohorts have been smaller than the 2021–2022 peaks — Winter 2026 funded 196 companies and Spring 2025 about 144, according to the WeAreFounders analysis of public batch data — while application volume is estimated at 20,000 or more per cycle. YC does not publish an acceptance rate; derived estimates for recent batches run from about 0.6% (Summer 2025) to roughly 1% (Winter 2026). The practical consequence is that a founder can now apply four times a year, and YC explicitly encourages reapplying with evidence of progress.

Everything YC tells you about the programme is consistent with how it filters applicants: speed, clarity and a willingness to build in public. Founders who want the wider context on how accelerators create investor access — demo day, signalling, alumni networks — should read the accelerators and demo days guide before deciding whether the 7% is worth it.

What are the YC deal terms and who is eligible?

YC invests $500,000 in every accepted company on two SAFEs: $125,000 for 7% on a post-money SAFE, and $375,000 on an uncapped SAFE with a most-favoured-nation clause that converts on the best terms you later offer another investor.

The standard deal page adds four details that matter for planning. The 7% is calculated including any existing option pool at conversion. YC takes a pro rata right to maintain ownership in later rounds. There are no milestones and no fees — the money is committed on the day of acceptance. And the company must be incorporated in the United States, Canada, the Cayman Islands or Singapore; companies incorporated elsewhere must flip into a parent in one of those jurisdictions, and YC introduces lawyers to do it. For a founder in Istanbul, Berlin or Bangalore that flip is a real cost and a real tax question, and it should be priced before applying.

The MFN SAFE is the piece founders most often misunderstand. If your next SAFE after YC is at a $15M post-money cap, YC’s $375,000 converts at that cap — the deal page’s own example gives 2.5% — so YC’s total ownership lands around 9.5% in that scenario, more if you raise at a lower cap and less if you raise higher. That makes the first post-YC round a negotiation with YC’s economics embedded in it, which is why the guide to reading a term sheet is worth revisiting before demo day rather than after.

Eligibility is otherwise minimal. YC accepts solo founders (though teams are preferred), pre-revenue companies, and founders of any nationality, and does not require a degree or a prior company. In practice the bar is a working product or a credible prototype plus a team that has shipped things together.

What does the YC form ask, and how should you answer each question?

The Fall 2026 form, as documented by third-party guides, asked for a 50-character company description, a 50-word answer to “what are you going to make”, 100–150-word founder bios, a story about a non-computer system you hacked, a 50-word “why are you doing this”, a one-line equity split and how long the founders have known each other.

The 50-character description

Name the customer segment and what they get, not the technology. “AI for logistics” fails; “Dispatch software for mid-size trucking fleets” passes. Reviewers read this line first and use it to decide whether to read the rest, and it is reused on YC’s internal lists and later on the Launch YC page if you get in.

What are you going to make?

Fifty words is nine or ten short sentences. Describe the product as it exists or will exist in three months, who uses it, what they do with it, and what changes for them. Avoid the vision paragraph; there is a separate field for why you are doing this.

Founder bios

Cite shipped work with links: a repository, a product page, a sales number, a paper. Titles and universities are weak evidence; a link to something you built that other people used is strong evidence. YC partners pattern-match for people who finish things, and the bio is where that pattern shows.

The hacked-system question

Tell it as a story with a constraint, an action and an outcome in about 100 words. The question exists to find resourcefulness; the worst answers are generic (“I negotiated a discount”), the best are specific and slightly uncomfortable.

Equity split and time known

Uneven splits are not disqualifying, but unexplained ones are. If one founder holds 80%, say why in the bio. A team that met two months ago is a risk flag; a team that has worked together for years is a signal. The data on this is summarised in the solo founder versus co-founder team analysis.

How does the YC video and interview work?

The video is one minute, all founders on camera, talking directly to the lens with no slides, music or edits; a screen recording of a video call is acceptable. The interview is ten minutes on Zoom with two or three partners who read the application as the call starts.

The video is a founder test, not a product demo. Partners are watching for who speaks, whether the founders agree on what the company is, and whether the energy is real. Say your names, what you are building in one sentence each, why you, and one concrete thing that has happened recently. Keep it under sixty seconds; guides that track YC applications report that over-length videos are a frequent and avoidable mistake.

The interview is fast and adversarial by design. Expect rapid questions on the market, the number you are proudest of, what you learned from the last ten customers, and why this will be big. Answer in one or two sentences, give the number, and let them ask the follow-up. Founders who rehearse a pitch lose time; founders who know their business cold and answer directly win. All founders must attend, and partners will direct questions at the quieter one.

Preparation for the interview is the same as preparation for any investor meeting, and the guide to what a strong pitch contains applies. The difference is compression: ten minutes means you must choose the three facts that prove the company, and lead with them.

How does Techstars select companies and what is the $220k deal?

Techstars runs three-month, mentorship-driven accelerators — 16 are listed on its site as of October 2026 — and since the fall 2025 batch invests $220,000 per company: $20,000 for 5% equity plus a $200,000 uncapped SAFE with an MFN clause, according to TechCrunch’s April 2025 report.

The terms are deliberately close to YC’s. TechCrunch’s example: if the next round values the company at $10M, the SAFE converts to about 2%, for roughly 7% total — YC provides more than twice the cash but takes more ownership. Techstars’ own accelerators page lists programmes spanning city-based (Boston, Boulder, Chicago, Columbus, London, New York, Tokyo), vertical (Space, Future of Food powered by Ecolab, AI Health Baltimore, Alabama EnergyTech, two healthcare programmes) and corporate-partnered (ABN AMRO Future of Finance, USC) formats, plus Techstars Anywhere, which runs remotely. Each programme has its own window and selection team; the London programme, for instance, opened applications on 24 August 2026, closes them on 18 November 2026, starts on 8 March 2027 and holds demo day on 3 June 2027.

Selection is run through a single portal (apply.techstars.com) and is more local than YC’s. Managing directors screen for programme fit — a space company applying to Techstars Space, a fintech to the ABN AMRO programme — and for coachability, because the model is built on mentor meetings in the first month. Techstars has said it selects hundreds of companies a year across programmes; per programme that means roughly ten to twelve companies from several hundred to a few thousand applications, an acceptance rate in the low single digits that is nonetheless several times higher than YC’s.

Techstars has restructured repeatedly since 2023, closing some city programmes, re-opening others under corporate sponsorship and changing its investment terms. Verify that the programme you want still exists and read its specific page for the current window before building a plan around it.

How does Antler’s residency model differ from YC and Techstars?

Antler invests in people before companies: it admits individuals — often without a co-founder or an idea — into a six-to-ten-week in-person residency, helps them form teams, then invests in the teams that pass an internal investment committee. Terms are set per country.

The application form is deliberately light: name, email, LinkedIn, stage (“idea to pre-seed” or “Series A to C” for the separate Elevate fund) and a city chosen from a list that spans Singapore, Jakarta, Bangalore, Tokyo, Seoul, Sydney, London, Amsterdam/Berlin, the Nordics, New York, San Francisco, Austin, Toronto, São Paulo, Dubai, Riyadh, Lagos and Nairobi. The real screening is a series of interviews that test problem-solving, drive and the ability to work with strangers, because the residency starts with people who do not yet have a team.

Terms vary by location and have moved in 2025–2026. For the Nordics, Antler’s July 2025 post described €200,000 split into €100,000 for 8.5% and €100,000 on an uncapped MFN SAFE, with up to €300,000 of matching into the first post-Antler round. For Singapore, the January 2026 announcement describes $100,000 for 10% via SAFE plus a $50,000 uncapped MFN SAFE, up to $250,000 of matching at 50% of institutional money raised, and a six-week in-person residency. The US page for Austin describes a typical initial commitment of $500K–$1M plus partner credits, residency of up to three months, no fee, and a requirement that at least one co-founder already hold US work authorisation.

Antler is the right route for a strong individual without a team; it is the wrong route for a formed team with traction, which should go directly to YC, Techstars or a pre-seed fund. A formed team can still apply, and Antler’s Nordic terms note that pre-formed teams can get investment decisions as early as week four.

What about Entrepreneur First, 500 Global and a16z speedrun?

Entrepreneur First is the closest comparator to Antler; 500 Global’s Flagship Accelerator and a16z speedrun are closer to YC but with different cash, fee and residency structures.

Entrepreneur First admits individuals in London, Bangalore and San Francisco, pays an equity-free stipend while they find a co-founder and an idea, invests up to $250K once a company forms, and commits up to $5M in follow-on through Series B, according to joinef.com. Applications go through apply.joinef.com; selected applicants attend a residential weekend before the matching process begins. EF screens individuals for outlier track records — people who have built products, organisations or startups and consistently outperformed peers.

500 Global‘s Flagship Accelerator in San Francisco runs four months (Batch 36: 30 November 2026 to 9 April 2027), gives up to $50K on acceptance with the possibility of up to $1M more over time, and charges a $35,000 Phase 1 programme fee, per the Flagship page. Founders must be in the US or planning to arrive within six months. a16z speedrun is twelve weeks in San Francisco with up to $1M — $500K for 10% upfront and up to $500K in the next round within 18 months — and, for sr008, a deadline of 1 November 2026, per a16z. Its ownership stake is higher than YC’s at the point of entry, but the follow-on commitment is larger and the firm’s growth-stage reach is unmatched; it suits consumer, games and AI-application founders who intend to raise from a16z later.

For a full list of programmes and funds with open routes beyond these five, including European and Asian options, see the global list of VC firms accepting applications.

How do the programmes compare side by side?

The table collects the terms, lengths, locations and windows as published by each programme in autumn 2026; the final column flags the thing each programme is really screening for.

Sources: programme pages and TechCrunch, reviewed October 2026. Deal terms change between cohorts — confirm on the programme’s own page.

Which programme should you choose, and what are the alternatives?

Choose YC if you have a shipped product and a complete team and want the strongest signalling; Techstars if you need a specific vertical or city network; Antler or EF if you are an individual without a co-founder; a16z speedrun if you are building in consumer, games or AI applications and want a16z’s growth capital behind you.

The alternatives are not inferior; they are different trades. A pre-seed fund with an open form — Hustle Fund, 1517 Fund, Seedcamp, Speedinvest — will negotiate terms rather than impose them and will not require a three-month relocation. Government and quasi-public programmes such as Germany’s HTGF, the EIC Accelerator or Turkey’s TÜBİTAK BiGG offer larger non-dilutive or low-dilution sums with slower processes. Both groups are covered in the Europe and UK guide and the Asia, Middle East and Turkey guide.

Whatever you choose, treat the application as a product. Every form in this article asks the same six questions in a different order, and the guide to what application forms screen for shows strong and weak answers to each. And remember that the programmes themselves tell you what they want: YC’s interview questions are published by former applicants, Techstars’ managing directors blog about what they look for, and Antler’s investment-committee criteria are discussed in its own posts. Read them before you write. Browse the full VCs Accepting Applications pillar for the regional lists.

Frequently Asked Questions