Insanity Inc.

That's where Brendan McGeever, 22, and Darin Weeks, 28, two more Penn graduates, are relaxing after working hard and playing hard at TU. As the beer, the margaritas, and the music flow around them, McGeever says, "I wasn't ready for college to end. None of us were. Now it doesn't have to."

TUers are living out their dream, Liemandt says. "They just came out of school, and they're like, 'I studied a lot. Now I want to work a lot. So don't bore me, and don't spoon-feed me. Give me really hard stuff and lots of responsibility, and I'll go deliver.' And by sheer force of will, they do deliver. They come up the learning curve very quickly."

That's the upside of youth. The downside is that these smart young people have virtually no business experience, and they often don't recognize the importance of teamwork. That's why each TU class is broken into small teams. Some teams do assigned projects. Others, like a group that is feverishly developing a car-order product for the Web, pursue their own initiatives. These new-product teams enjoy Liemandt's blessing -- for now. "They are totally clueless at times," Liemandt says. "They come up with an idea for a product, and they don't let anything get in their way, even though their basic business plan has this huge hole in it. They expect magic to occur. So I tell them, 'You've got to explain the magic to me.' "

Stars and Sponsors

Trilogy didn't get really serious about the people factor in its business until five years after its founding. Back in 1994, during a review of the latest crop of recruits, a Trilogy vice president mentioned that one hot prospect had gone elsewhere.

"What happened?" Liemandt asked.

"I don't know. It's not really a big deal," the VP replied.

But to Liemandt, it was a huge deal. " 'We're giving lip service to how we build our company,' " he recalls thinking. He asked John Price, 39, vice president of marketing, to give the issue some thought. Price's solution was to create Trilogy University. Price would hire the 50 most promising college graduates that he could find, and the CEO would teach them everything they needed to know about developing and marketing software.

"Don't I have another job to do?" Liemandt protested. But he knew that Price was right.

The idea behind TU was to re-create the spirit of Trilogy's startup years -- indeed, to turn Trilogy into a perpetual startup. The logic is clear: Ramp up recruits, overload them with responsibility, let them suffer a bit, and see what they come up with. Ultimately, at TU, results are what get rewarded.

That results-oriented logic applies to the other half of the recruiting equation: assessing the candidates who want to join the company. Trilogy doesn't just evaluate its recruits -- it also evaluates the evaluators. "There was a lot of disagreement over candidates," says Liemandt, "so we said, 'Let's track the data and see who's right.' "

In job evaluations, Trilogians are ranked on a scale of one to three, with one being a "star." Employees whose recruits are on their way to becoming stars become "sponsors." Today there are 60 such sponsors in the company, and rarely does a candidate get hired without one. In sponsoring a candidate, Trilogians are not only saying they believe that he or she will be a star performer; they are also accepting responsibility for his or her performance. "If you hire someone and you're wrong, it's your job to be a mentor and to fix the situation," says Liemandt.

To improve the process of evaluating candidates, Trilogy turned another business convention -- "only managers interview recruits" -- on its head. To a large degree, Trilogy's best interviewers are also its best developers and programmers, its best consultants and salespeople. After all, who better to recognize talent than the company's most talented in-the-trenches employees?

Graham Hesselroth, 26, one of Trilogy's top developers, conducted about 350 interviews last year. He's considered to be one of Trilogy's toughest interviewers and one of its best judges of talent: He sponsors only a handful of recruits, but those whom he does sponsor are virtually assured of becoming stars. His approach is direct: "You'd better blow me away." If Trilogy is going to keep improving, he reasons, it needs to hire developers and consultants who are better than the ones now on board.

Hesselroth, who wears his hair in a long ponytail, seems more relaxed than other Trilogians, most of whom radiate a caffeine-enhanced energy. Calmly and methodically, he probes a recruit's mind, testing her degree of expertise, her passion, and her ability to solve multiple problems under pressure. He asks himself, "Would I trust this person to write a critical piece of code for me on my next project? Will I get along with her? Is she going to go postal on me during the next crunch cycle?" He switches back and forth between problems, trying to confuse the candidate, to overload her mental circuits: "In an environment like Trilogy's," Hesselroth explains, "you need the capability to characterize complex systems and to develop insights in a short amount of time."