YC's post-money SAFE: where the standard contract diverges from the Brazilian cap table

Anyone negotiating funding with a US fund today has probably already received this document in their inbox: Y Combinator's post-money SAFE, available on YC's official documents page (ycombinator.com/documents). It's short, standardized, and, according to YC itself, the instrument it uses to fund its own program startups. The problem isn't what the contract says. It's what it assumes about where the company is incorporated.

How the equity math actually closes

The post-money SAFE inverts the logic of the pitch: instead of negotiating valuation first, the founder chooses how much to raise and what slice they're willing to sell, and the valuation cap comes out of that math. YC itself gives the example: raising $1 million while selling 15% of equity implies a post-money cap of approximately $6.7 million ($1 million divided by 15%). Raising $500,000 at the same cap sells about 7.5%; $800,000 sells about 12%.

You can stack different caps and arrive at the same result: $500,000 at a $5.5 million cap (~9%) plus $500,000 at an $8.3 million cap (~6%) add up to the same ~15% sold. This transparency is YC's central argument for the post-money SAFE: the founder knows exactly how much they're selling at the moment of signing, without waiting for the priced round to find out the real dilution.

The discount nobody notices: the post-round option pool

The fine print that most often catches founders off guard sits in a single sentence in YC's own FAQ about what the post-money cap includes. The cap is "post" all the money raised in SAFEs and "post" the option pool that already existed before the priced round. But it is NOT "post" the new or expanded option pool that the company creates as part of that same round.

In practice, this means that if the founder expands the option pool at the time of the priced round (which investors almost always request), that additional dilution falls on founders and already-converted SAFE holders; it is not absorbed by the cap. It's a real mechanic of the instrument, documented by YC itself, and it's exactly the kind of math that separates those who close the round at the equity they planned from those who only discover the real dilution at the cap table after signing.

MFN: the clause that rewards whoever signs last

The SAFE with an MFN (Most Favored Nation) clause has no cap and no discount: it automatically inherits the best terms of any SAFE the startup issues afterward. YC is specific about which date counts for that inheritance, and it's a detail that changes the order in which the founder should close investors.

The date the MFN SAFE uses is the later of signing the SAFE or receiving the wire, not when you sent it. So close your lower valuation cap SAFEs (fully wired) before you issue an uncapped MFN. You sign last and control the date.

Y Combinator, official SAFE guide

Translating this for anyone in the middle of a round: an MFN investor signed before the others may end up inheriting a worse (higher) cap than the founder intended, if a lower-cap SAFE is only signed and funded afterward. The closing order isn't bureaucracy; it's the difference between the MFN inheriting the $5.5 million cap or the $8.3 million cap from the earlier example.

Pro-rata lives outside the contract

Pro-rata rights, which give the investor the option (not the obligation) to invest in the next round to maintain their ownership percentage, are not in the SAFE. YC makes this explicit: pro-rata lives in a standardized, optional side letter, separate from the main contract.

This gives the founder real negotiating room. YC's own guide suggests evaluating it case by case: an investor putting in $500,000 has a reasonable argument for requesting pro-rata; one putting in $10,000, not necessarily. It's a point of leverage that gets lost when the founder treats the document package as a closed package and signs the side letter without questioning it.

Why Brazil doesn't appear on YC's list of countries

Here's the divergence that matters most for anyone in Brazil. In the downloadable forms section, YC offers three SAFE versions for US companies and, separately, valuation-cap versions for companies formed in Canada, the Cayman Islands, and Singapore. Brazil is on neither list.

This isn't an irrelevant detail. It means there is no official version of YC's post-money SAFE designed for a company incorporated in Brazil under the Lei das S.A. (Brazil's corporate law governing joint-stock companies) or as a limitada (Brazil's limited-liability company structure). Anyone in that position has, in practice, three paths:

• Do a corporate flip into a Delaware C-Corp (the most common path among Brazilian startups that go through US accelerators) and use the American SAFE without adaptation.
• Structure a holding company in Cayman or Singapore, using the non-US version of the form with the support of local counsel, as YC itself recommends.
• Negotiate an adaptation of the instrument to the Brazilian corporate structure, usually via a convertible loan agreement (mútuo conversível) or an instrument provided for under Lei Complementar 182/2021 (Brazil's Legal Framework for Startups), which has a conversion mechanic different from the automatic one set out in the SAFE.

The difference isn't just naming. A traditional Brazilian convertible loan carries the debt features that YC itself lists as the reason the SAFE isn't a convertible note: accruing interest and a maturity date that forces repayment or conversion. This changes the founder's position in a future valuation negotiation, because the Brazilian instrument can create a cash obligation that the SAFE, by design, never creates.

The counterpoint: the flip isn't a bug, it's standard practice

It's worth acknowledging the best argument against treating this absence as a serious problem: for startups aiming at US capital and a YC batch, flipping to Delaware has been routine for years, not a new barrier. American venture funds prefer investing in a Delaware C-Corp for tax and governance reasons that predate the SAFE, and a good share of Brazilian startups that raise from US funds are already born into, or migrate to, that structure before they even think about SAFE terms.

In that sense, Brazil's absence from YC's list of countries isn't an unexpected gap: it reflects a corporate decision the founder has probably already made (or will have to make) when targeting this kind of investor. The real risk isn't the missing form; it's the founder assuming they can simply sign the American PDF with a Brazilian company behind it, without realizing that the SAFE's automatic conversion mechanic depends on a corporate regime (Delaware's) that Brazil's Civil Code and Lei das S.A. don't replicate in the same way.

What this means for anyone closing a round right now

Three concrete points are worth checking before signing:

• Confirm the corporate structure before the cap. If the investor is sending YC's standard post-money SAFE, they're assuming (or requiring) a Delaware C-Corp behind it. Aligning this with counsel before discussing the cap number avoids redoing the paperwork.
• Redo the dilution math including the new option pool. The post-money cap doesn't absorb the pool expansion that happens at the following priced round; that dilutes the founder and already-converted SAFE holders, not just the new investors.
• Negotiate the signing order of SAFEs with an MFN clause. Closing the SAFEs with a fixed cap first, and only then the MFN one, prevents the uncapped investor from inheriting a worse number than intended.

YC also offers its own tool, "Send a SAFE," and a CLI designed to be operated by AI agents, capable of generating, signing, and sending the contract in about two minutes. It's a sign of how much the instrument has become a commodity on the American side. For the Brazilian founder, the tool's speed doesn't answer the question that comes before it: in which jurisdiction the company that's going to sign this contract actually exists.