Government bureaucracies and start-ups are usually two things that don’t mix well together, but the National Science Foundation is looking to change that with a $10 million grant for an start-up education program intended to teach the nation’s top scientists and engineers how to become entrepreneurs.

The Innovation Corps program — which starts in September at Stanford University –will give $50,000 to 100 different teams (3 or more people per team) every year to go through an intensive entrepreneurial education class. The I-Corps class will be modeled on a Stanford engineering class called Lean LaunchPad that was taught earlier this year by serial entrepreneur Steve Blank and a coterie of entrepreneurial thought leaders, technologists and venture capitalists. The grant, which runs to $10 million, covers five years of operations of the program, which will be taught quarterly to 25 teams at a time.

And unlike angel funding or investments from tech incubators such as TechStars and YCombinator, the teams don’t have to give up any equity to the government or to Stanford Technology Ventures Program (STVP), which is handling the education portion of I-Corps.

The point?

To change how engineers are trained in the U.S. so that they can go create innovative businesses that grow the economy and create jobs.

“In meeting the country’s needs for innovators, we want to develop students who can work in and manage creative teams; communicate effectively and think critically; understand business basics; and be leaders who can solve open-ended problems,” said NSF program director for Education and Human Resources Don Millard.

But don’t think of it as a semester-long MBA school.

Think of it as a recognition by the government that entrepreneurship isn’t being taught by MBA programs, but instead has been developed collectively by the culture of Silicon Valley, which has developed a culture of fast development where products are continually upgraded in small — often daily — steps, constant attention to customer needs, and a focus on measuring everything.

“The U.S. government is doing YCombinator,” said Steve Blank, referring to the wildly successful tech incubation program started by Paul Graham and Jessica Livingston in 2005, which gives a small amount of cash — along with advice and pressure — to coders with ideas, pushing them to create companies in months. YCombinator alumni include Dropbox, Reddit (purchased by Wired.com’s parent company), Loopt, AirBNB, Scribd and Heroku.

Teams selected via the National Science Foundation will take a class at Stanford that centers on the search for a business model by having students continually test their ideas, products and assumptions against the real world and refine or switch completely in response (see the curriculum).These are processes known in Valley lingo as customer development, agile development, and pivoting.

What that means, according to Blank, is that he and the Valley have learned how to teach entrepreneurship to non-business majors with a methodology that can be replicated — which he credits the NSF for recognizing.

“The NSF was smarter than I was,” Blank says. “They told me, ‘You have turned finding a business model into nothing more than the scientific method.’”

Blank is co-teaching the class with Stanford entrepreneurship professor Tom Byers and venture capitalist Jon Feiber. Blank’s also quick to give credit to others, including former student Eric Ries — known for pushing the idea of “Lean Startups” — and Alexander Osterwalder, the author of a book called Business Model Generation. (Osterwalder will be guest teaching the NSF grantees this fall, as will John Burke from the VC firm True Ventures.)

But how is the program any different from an MBA?

MBA schools, in Blank’s view, are about teaching people how to execute a known business model, which is why it’s a masters of business administration — essentially focusing on how to run a division of GE or GM. His class, and by extension the startup culture in Silicon Valley, is is about creating another GE, by searching for new business models.

“In the 20th century, business innovation was centered on Harvard business school,” Blank said. “I will contend the center is now on the West Coast and it’s not in business schools.”

“You can’t even describe to people who don’t live in this culture how fast innovation happens,” Blank said. “You go on vacation in Silicon Valley, and there’s an entire new industry created by the time you get back.”

Blank’s spring class — known as the Lean LaunchPad — forced teams of engineering students to start a company together and rapidly test and adjust their ideas by testing it against the real world. One group started with the idea of creating an autonomous grass cutting machine that could be used in parks, along highways and on golf courses. But after testing the hypothesis and finding customers uninterested, the group switched over to the idea of creating a robotic weed-killing machine that could distinguish between crops and invaders and zap the latter with lasers.

The nice thing about the class, according to Blank, is how much of the work lands on students, who have to continually give presentations to classmates — which led students to often spend more than 40 hours a week on their “start-up,” even as they were taking a full-load of classes.

“The key is to make the students work their ass off, not me,” Blank said. “The peer pressure of having to present embarrasses them. You make the process pretty easy, but the work is hard.”

Blank has already been laying plans to expand the LaunchPad idea beyond Stanford, taking it to UC Berkeley and Princeton — though he’s not going to be teaching all of those.

“The process is eminently cloneable,” Blank says.

So how do students get to join the program? Well, that involves a application process and only those who have had an active science and engineering award from the NSF in the last five years are eligible.

That process will be handled by the NSF, which as an organization that used to handing out billions a year in research grants, is well-suited to vetting hundreds or thousands of applications.

“I think it’s a big deal for the U.S.,” Blank says. “It takes the best we have got in engineering and science and the goal is not just research — it’s about whether we can build companies and create jobs.”

Photo: Steve Blank courtesy Steve Blank.