Jason Njoku doesn’t like being anyone’s bitch.

As I learned the last time I saw him– over machetes in Nigeria– that includes vigilante mobs. But that also includes YouTube, Amazon Web Services, and US-based video ad networks. And those are the people causing him the most problems since we last spoke.

Njoku is building IrokoTV. It’s not only one of the most exciting Web companies to come out of Africa, and it’s one of the most interesting original video startups I’ve seen in the world. He’s painstakingly building a modern distribution engine for the second largest movie industry in the world by volume: Nollywood, or Nigerian cinema. Brick. By. Brick.

Iroko does it all: They roam the markets of Alaba cutting deals with independent film producers, they scrub the content for remove copyrighted content, they subtitle it, they enter all the metadata, they convert it from outdated formats, they enter all the relevant information in a IMDB-like database, and upload it for hundreds of millions of people in the Nigerian diaspora to rabidly consume. (That’s him haggling with producers above; Njoku is on the right.)

Until a few months ago, he let YouTube do the rest. He started a turn-key YouTube channel, that handled the cost and challenges of serving up gigabits of video every month. They handled ad sales. And even some distribution, putting Nollywood movies in some genre lists.

When I last saw Njoku in Nigeria that was working well. He was doing eight million streams a month, and on a $1 million revenue run-rate. But since then, his traffic has soared, and the relationship soured. In Njoku’s view, YouTube has become so maniacal about pleasing Hollywood and doing original, high-level US-centric programming that they just didn’t care about what someone like Iroko was doing.

Iroko ended the year with 150 million views, and is poised to do 250 million views this year. That’s not nearly as large as other channels, but bear in mind, a view for Iroko is typically a full-length movie. And these are still early days for the company. The bulk of that traffic is coming from the US and the UK, as broadband is in its infancy in Africa. Nigeria alone is a 150 million person market.

Ultimately, Njoku felt like he was hustling on the streets of Nigeria to cut content deals and bring the world content it couldn’t see any other way. He was holding up his end of the partnership, but YouTube couldn’t be bothered to help him in the slightest. “They just didn’t understand our content,” he says.

One example: To make money, he had to put in ads every 15 minutes which greatly interrupted the flow of a full-length movie. He spent months asking YouTube if he could cluster several ads at the beginning the way other channels on YouTube do. They simply didn’t answer or refused to budge, saying those were special arrangements for other partners– even though some of those partners had lower volume.

And Google was causing him other problems too. Or more to the point: His location in Nigeria was causing him problems with Google. Because of the country’s widespread problem with 419 scammers, large Internet companies will routinely cancel accounts that have a Nigerian IP address. He was trying to pay Google thousands of dollars a day for paid search keywords, only to have his account routinely wiped clean. (His answer? Spending more to advertise on Facebook, where his location didn’t pose a problem.)

He could, of course, do what many Nigerian entrepreneurs do, and hide his digital location, saying he was headquartered in Ghana. He does indeed have an office in London and is in the process of opening one in New York.

But no. Screw that. Njoku is building a company exporting the raw, creative gold of Nigeria’s film industry. He is going to do this on his terms, as a Nigerian company. And it became increasingly clear that he was going to have to build the company without YouTube for that to happen. “We have so much content and people love it and I’m paying for it to put it on YouTube for free,” he says. “It was just clear we were never going to be a priority. That’s the downside of building a business on someone else’s platform. I was just sick of being YouTube’s bitch.”

So earlier this year, Njoku  did something bold. He decided to close his YouTube channel that was paying him millions of dollars in an attempt to build something even bigger. (He’s in the process of migrating users and content now, so it’s still up for a bit longer.) He opened a stand alone site, using Amazon Web Services, Ooyala and several video ad networks.

It hasn’t been easy. And there are days he regrets the decision. On YouTube he just had to be popular, and he’d make money. And being popular was easy for Iroko.

On his own, the traffic is still growing, but he has “nosebleed expensive” costs and revenues are harder to come by. And he’s run into many of the same problems being a Web company in Nigeria. He had to badger Ooyala into taking him seriously, flying to New York and showing up at their offices before a salesman would meet with him. He paid them a quarter of a million dollars up front to prove he was serious. When Iroko zoomed to being one of Ooyala’s top 20 sites, things changed rapidly. To Ooyala’s credit, once they saw the demand for Nollywood films, they’ve bent over backwards to make life easier for Njoku, unlike YouTube.

When he tried to sign up for Amazon Web Services, they insisted he didn’t know the difference between “gigabytes” and “megabytes” when he said the site was doing 800,000 gigabytes per month. They simply couldn’t believe how many views his channel was getting and how many of them were for full-length movies. He had to get Ooyala to send the numbers over before AWS would take him seriously.

And as Njoku has gotten more serious about going after Nollywood pirates– with a seven person legal team in the UK and US– they fight back by filing counterclaims against him. Even though he has copious documentation on paper and in video of producers verifying they’ve sold him the digital license, the confusion has caused Amazon to shut down his site numerous times for seemingly no reason, just like AdWords did. That costs him views and money each time.

He even got rejected when he tried to enter his company information into CrunchBase– never mind I written about his company on TechCrunch twice, amassing thousands of social media Likes and Tweets and follow-on press calls from several major networks and newspapers.

“Hey ho,” he says, brushing it off. He’s more convinced than ever he’s onto something huge and all of these problems just mean his competitive moat is that much wider, deeper and filled with more metaphorical alligators. “My people– wherever they are in the world– love what we’ve done. As long as I have that, we’re going to be OK,” he says.

Njoku told me all of this over a fancy San Francisco dinner last night. It was the least I could do to thank him for bribing my family out of hot water last summer. It was a markedly different setting than our last interview. I was far more relaxed. I wasn’t sweating in 100 degree Nigerian heat.  I was not five months pregnant. And the sharpest knife in the room was for slicing seared scallops into bite-sized pieces.

But Njoku was the same: Friendly and unflappable, with an easy smile but a clear take-no-bull-shit determination behind it. “I’m not going to hide who we are,” he said. “We have 71 people working for us in a country with 50% unemployment. We did $1.3 million in revenues last year. We are the largest Internet company in Lagos by nearly any definition. We want to be Nigeria’s Netscape moment.”

There’s more he told me about Iroko I promised him I wouldn’t report…yet. But let me say to those doubters: I’ve seen Njoku in action in Nigeria. I’ve been to the markets with him. I’ve watched producers descend on him, hungry for his ability to get their work into the world. And on the other end of the broadband pipe, I’ve seen viewers hungry to consume that content.

I can attest that this Nigerian entrepreneur is not only a legitimate businessman, he is one of the most hardcore entrepreneurs I have ever met.