Last week Naval Ravikant went to an industry dinner. He asked a friend in the venture business how things were going. The friend slumped over in his chair, shrugged sadly and said, “The business is becoming commoditized.”

It’s an extreme interpretation, and not everyone shares it. Times have never been better for a handful of firms who are rolling in the returns, raising as much from LPs as they want and still doing business the way they always have.

But matters have also never been more polarized for the VC-haves and have-nots, and this sad-sack VC has a few people to blame. Chief among them is his friend Ravikant, whose site AngelList has dragged the stealthy, back-room world of venture capital kicking and screaming into the light — something many industry watchers never thought could be done.

And now, AngelList is doing the same thing it did to VCs to recruiters. But we’ll get to that in a minute.

People have wanted to dismiss the potential of AngelList ever since it launched. At first, people said only loser companies would want to share data and telegraph fundraising plans so publicly. Then VCs said they’d never invest in something that didn’t come through traditional channels. Other VCs said they’d never do something as banal as leaving “comments” on a company’s profile. They’d have those conversations privately or not at all.

But the AngelList juggernaut has steamrolled right through those objections. And now, it’s not even just about seed deals. Yesterday, we wrote about the first large Series A completed on the site, when Rally raised a remarkable $7.9 million round.

The important thing about the Rally deal: This was a Valley company with two big leads in Floodgate and Greylock. It could have done the round the traditional route, but chose AngelList instead. This is the thing: AngelList is not about everyone being able to raise money or everyone being able to call themselves an angel. It’s about the very best companies actually choosing to operate out in an open platform. The image that so many people hoped for — that AngelList would only be the trolling ground for the desperate, for the unconnected — just hasn’t proven true.

And that’s exactly what some people find so scary about it.

In the case of Rally, Ravikant argues the referrals from AngelList converted at a much higher rate than emails existing investors sent, because the social niceties were stripped out of the transaction. Put another way: If an investor on AngelList signals interest, they are truly interested. If a friend of Mike Maples says, “Ok, I’ll take a look, Mike,” he may just be being polite.

What’s scary about AngelList is it strikes at the heart of what VCs have held dear for decades: Proprietary deal flow. “The Holy Grail is finding a company before anyone else,” Ravikant says. “That is disappearing. The idea that you are the only investor to see this deal is highly, highly, highly, highly unlikely.” (That’s four highlys, if anyone was counting.)

It’s hard to overstate how terrifying that idea is for the Valley establishment. Partners are hired because they have proprietary deal flow. VCs sell that to LPs as the reason to invest in them, not the VC next door. The desire for more of it is why we recently discovered that venture firms were paying entrepreneurs to secretly invest on their behalf.

Everyone wants to know about the hot new deal before anyone else. That’s how you make your money in the venture business. And AngelList and Ravikant are standing on a hill screaming that the Web, tech blogs, and sophisticated entrepreneurs have destroyed the idea of proprietary deal flow forever. Pour the sad sack VC at dinner another drink. For most of the Sand Hill Road, that is very, very bad news. And, whether fans of AngelList or not, most will still tell you it’s flat out not true, that deal flow is still very much alive and kicking.

Still, you can’t deny AngelList’s burgeoning power. While uncomfortable and threatening for some, it has caused a full-on stampede on the site. We use AngelList to provide widgets at the end of our stories, and it’s hard to find a new startup, no matter how nascent, that doesn’t have a profile on the site already. Globally, angels are fighting to get listed.

It’s significantly harder to invest via AngelList than it was to buy secondary shares in Facebook. Not only do you have to be an accredited investor, but AngelList has to verify that at least two investments you’ve made were pure cash investments. Someone else in the community has to approve you. And even then, you are on probationary status. There are 3,500 angels on the site, which sounds like a lot, until you realize there are more than 20,000 applications in the system.

Here’s the thing: Those 20,000 likely aren’t getting in. At least not any time soon. “We stopped going through the inbox,” Ravikant confesses. It’s just too hard to process manually, and they are working on a better way to process the ever-growing queue of hopeful names. “The only way someone gets ahead now is if someone I know emails me and says they should be added as an investor,” he says.

The caution signals that even Ravikant is a little scared of the power of what he has built. When we talked about the millions in cash thrown at Rally over his platform, he could have strutted around bragging about the power of AngelList, lobbing smug told-you-so’s at the haters. He didn’t. He repeated this several times instead: “I hope they make their investors money.” That’s how he knows he’s really done something here.

Ravikant emphasizes repeatedly that AngelList isn’t a king maker. At best, it’s a king-enabler. It can only help great companies; it can’t make a shitty company great. “We get 100 company profiles created every day,” he says. “There aren’t 100 companies that get funded every day. Just like YC and Harvard, the 1 percent are the ones that do well. The next 2 to 3 percent do okay. We can’t make an unfinanceable company financeable.”

“This is the biggest misconception,” he continues. “People put up this tiny, little profile saying they’ll be the best food recipe site on the planet, with no data or information and later they’re complaining on Twitter that AngelList didn’t work. You have to put a lot of information up there; a lot of companies aren’t comfortable with that.”

Ravikant takes who does what on his platform very seriously. He doesn’t want it to be an example of some wild 2010s-era exuberance that people point to a decade from now, the way everyone mocked Draper Fisher Jurvetson’s failed “meVC” fund after the dot com crash. He doesn’t want this to end in tears. He doesn’t want this to be a flash in the pan. He wants this to change the way the venture business works, for the better, forever.

“This is fun, but it is moving very, very fast,” Ravikant says. “VCs invest over four year cycles with ten year vesting cycles. Look at how much the business has changed in the last three to four years.”

He tells me about a startup who is raising money now, with a flood of recently minted Facebook money anchoring the deal. “When they told me the price, I almost fell out of my chair,” he says. “There’s a lot of recycling of capital going on. There’s a lot of madness in this market.”

Ravikant has long said he has a big vision for AngelList, but won’t talk much more about what that is. But eagle-eyed startups may have noticed another part recently: a talent vertical. Go here to see it. Rather than operating like a normal job board, it matches people who failed to raise money or whose startup didn’t work out with companies that have yet to raise money.

It’s an ingenious approach, and true to AngelList’s savvy use of technology to shortcut real-world inefficiencies. Most startups who can’t find talent complain that too many people are starting their own companies, and that drains the market of would be CTOs and VPs. AngelList finds the ones that don’t quite work and efficiently feeds that talent back into the system.

And it seems to be working.

These days, AngelList is facilitating some 200 mutual introductions per week between startups and investors. It’s taken years to get to that velocity. But in just the last two weeks, the talent vertical — which AngelList never announced and almost no one has written about — is doing 600 mutual introductions a week between talent and startups looking for talent. “It’s already three times the velocity of the fundraising side of the company,” Ravikant says.

Holy. Shit. If you thought finding funding was a challenge for startups, AngelList has struck at the motherload of startup angst by taking on the war for talent. This could cause an even bigger impact in the startup world.

Of course, those are my words, not Ravikant’s. He shrugs, “We’re just a tool in the tool belt. Hopefully the best. But just one tool.”

(I should disclose that I met Ravikant, when PandoDaily first launched, to talk about partnering up to provide widgets on our stories. There’s no financial relationship between the companies, and there’s no information they give us that you can’t get on the site itself. But we clicked immediately, because we were both underdogs trying to bring this industry into the light, and both have a very long term view. He’s agreed to be our guest at PandoMonthly in November where he’s promised to share much more about AngelList’s big vision. It should be a good one.)

[Illustration by Hallie Bateman]