Last week, at a conference in San Francisco, I was asked to interview Intel Capital’s president Arvind Sodhani, who also holds the title of executive VP of Intel Corporation. The idea was to give the audience insight into how the nine-year-old corporate venture unit – which employs 85 investors around the world and invests between $300 million and $500 million each year – does what it does.

Suffice it to say that it’s complicated. Here’s a cheat sheet, though, for investors and entrepreneurs looking for more, ahem, intel.

Intel, like most corporate VCs, won’t invest in something that looks like it could be a financial home run but has no bearing on the company’s business. Every investment has to have the potential of delivering both a strategic and financial return. The good news: In one form or another, Intel’s business touches almost every sector out there, so it’s investing in everything from wearables to semiconductors to Hadoop, the bedrock software of so-called big data businesses — which Intel sees as a growth sector. Indeed, readers might recall that in late March, Intel forked over $740 million for an 18 percent stake in Cloudera, which produces the most popular version of the Hadoop software framework.

Which raises another point: Intel Capital is stage agnostic. While it sometimes makes a swing-for-the-fences deal like Cloudera, it’s also willing to fund very nascent ideas, Sodhani told me last week, calling the organization’s “sweet spot between $5 million and $20 million – that’s where we’re doing the bulk of our investments.”

As for where it’s making its bets geographically, Sodhani said that half of Intel Capital’s investments are here in the U.S., with the rest distributed globally, including in China, where the organization recently committed to invest up to $100 million in companies working on smart devices, as well as Israel, Russia, India, Brazil, Japan and elsewhere.

Unsurprisingly, some places work out better than others. Intel Capital has backed companies in places like Vietnam and Chile, for example, but hasn’t been able to consistently find deal flow in either country. It also recently placed an investor in Nigeria to scout out opportunities in Africa, though Sodhani said it “takes a year-and-a-half to two years before someone new in a country can get going and see investments . . . When you arrive in a new country, it takes time to figure out the legal framework, what instruments are available . . . there are lots of different issues.”

One of them is helping to develop a tech-friendly ecosystem, which Sodhani credits Intel Capital with doing in a variety of places like Vietnam, where startups are still a relatively new phenomenon. While there are plenty of founders, said Sodhani, wresting potential employees out of their solid jobs to work for those founders is still in an upward battle. “Entrepreneurs are willing to take the risk; the hard part is how do you get the rest of the people to come and join the company.”

Before we’d parted ways, I’d asked Sodhani to share how investment decisions are made within Intel Capital. He suggested that, despite the sprawling design of the organization, it isn’t unlike most venture firms. Every Tuesday, it holds a weekly partner meeting where all deals receive some air time and potential new investments are presented. Company experts are then brought in to discuss and evaluate new funding prospects, questions are asked and researched, and after at least a second look at a company, a committee of five people within Intel Capital decides whether or not to back it. (Sodhani says that Intel Capital can “move lightening fast when there’s a great deal if we’re made aware that it’s a competitive situation. We can put together a term sheet in less than 24 hours if we want.”)

Like a lot of firms with deep pockets, Intel Capital is also willing to overlook price if the technology is deemed as a must-have. “Valuations are very [high], and I’d say that we’re probably getting to a point where we need to be careful of them,” Sodhani said. Still, he’d added, “some things we have to hold our noses and say, [Let’s move forward], because the technology is important to us.”

Open source software is one of those things, he continued. “It’s becoming very expensive, but open source is producing lot of software that’s critical. The whole trend of IT migration to the cloud — that’s a $30 trillion idea.”