(FORTUNE Magazine) – REMEMBER that pudgy kid in the eighth grade, the one who liked to concoct bombs in his mom's kitchen? Now he's one of America's premier software designers. Or that other brat, the one who got in trouble for taking apart his dad's lawn mower? He's drawing up plans for the next generation of electric cars. The R&D divisions of most FORTUNE 500 companies are full of people like these. The mischievous tinkerers who seemed so annoying in grade school have grown up to be America's hottest engineers and programmers. Whether managers understand techies or not, they have to treat them right because a lot of American competitiveness depends on them. More and more, the battle for global market share is fought in corporate R&D labs, and engineers are the front-line troops. Japanese and European companies have been gaining ground in high-tech fields partly because they honor engineers. In Japan engineers are well represented in top management: Two- thirds of the CEOs of large Japanese manufacturing companies have engineering or science backgrounds. In the U.S. only one-third of such CEOs have technical expertise. Says Yoshi Tsurumi, a professor of international business at New York City's Baruch College: ''The company usually reflects what the CEO knows, and most American CEOs don't understand engineers. Japanese firms look for technical solutions to their problems, while American firms look for financial and legal solutions.'' Whatever the troubles of American R&D, a shortage of bright young engineers isn't one of them. The field surged in popularity in the late 1970s and early 1980s, spurred by increased defense spending and the birth of the personal computer. Today the U.S. has 1.8 million engineers, up 29% from ten years ago; their median age is 39. Although the number of women engineers has more than doubled over the past decade, the profession is still 92% male. It's also 89% white, despite recent inroads by Asian Americans and foreign nationals (notably from India, Taiwan, and South Korea). About half of all engineers go on for a master's degree or doctorate, but advanced education isn't always required for success. Some of the best programmers at Microsoft, for example, are college dropouts -- including billionaire CEO Bill Gates, who quit Harvard after his sophomore year. Here's the biggest surprise: Very few engineers are nerds. Not many wear those plastic pocket protectors on their shirts. Although their love for doohickeys and gadgets sometimes borders on the obsessional, they don't shy away from human contact. Engineers are, by and large, very social animals, but managing them is different from managing salesmen or accountants or assembly- line workers. Of the three dozen engineers and programmers interviewed for this story, not one mentioned career advancement as a primary goal. A typical comment: ''I don't really have a career plan'' -- a statement that would horrify many corporate ladder-climbers. What does motivate engineers? Technical challenge. They attack problems like crossword-puzzle addicts, racking their brains until a solution emerges. Their chief reward is the work itself. Samuel Florman, a civil engineer who has written several books about his profession, coined a phrase to describe its satisfactions: ''the existential pleasures of engineering.'' How can a company harness this creative energy? How can it motivate people who are more or less indifferent to raises or promotions? The first step is to find managers who are engineers themselves, or at least understand the engineers' work. A manager in the semiconductor industry, for example, must know a lot about electrical engineering in order to discuss the prospects of a new microprocessor with the R&D staff. Says Neil Norman, a past president of the National Society of Professional Engineers: ''If you're doing state-of- the-art work, the top managers need some technical expertise. Otherwise they're going to get blindsided.'' It's no accident that so many companies with superlative R&D -- Motorola, GE, and Intel, for example -- have CEOs with engineering degrees. Managers also need to infuse the engineering staff with the kind of can-do spirit that pervaded NASA's Apollo moonshot program during the 1960s or the Manhattan Project during World War II. Engineers must be assured that management thinks what they're working on is important and supports their work wholeheartedly. At companies like Microsoft and Hewlett-Packard, the CEOs wander through the hallways of their R&D divisions, spurring on the engineers and occasionally grilling them about their progress. Neil Konzen, a software developer at Microsoft, has been in on many lively review meetings with CEO Bill Gates. Says Konzen: ''Bill doesn't just sit there and say, 'Great job, great job.' He'll yell at you about the state of your project. The guy knows what the hell he's talking about, so he's able to find the weak spots. We always come out of those meetings with something we have to change.'' Companies too often segregate engineers into R&D ghettos, leaving them no one to talk to but each other. That's bad for the engineers and for the company. When engineers talk to salespeople and customers they get valuable insight into the market for their products, which improves their designs. Managers can encourage interaction by assigning engineers temporarily to marketing and customer service departments. Software designers at Microsoft regularly listen in on customer calls to the company's product support group. Pat Gelsinger, an engineer at Intel who helped design the hot-selling 486 microprocessor, says there is a growing demand for ''Renaissance engineers'' who have good business sense in addition to their technical knowledge. An R&D program doesn't have to be big to be successful. Arco, the petroleum giant, in 1985 dismantled its 500-person corporate R&D lab near Chicago on the theory that the industry was too mature to require much new research. Three years later, when it became clear that California environmental regulators would adopt drastic new standards for auto emissions, Arco hastily assembled an ad hoc research team to create a clean-burning gasoline that would be economical to produce. HEADING Arco's clean fuels group was Dan Townsend, 42, a chemical engineer with a B.S. from the Colorado School of Mines who had moved up through the manufacturing department. Townsend's background was mainly in process engineering; in the petroleum industry that means designing new refineries for optimum performance. His distinguishing trait, like that of many engineers, is an infectious enthusiasm for his work. After getting a go-ahead from Arco's senior managers, Townsend organized an eclectic team of two chemical engineers, one mechanical engineer, and three chemists. The oldest had more than 20 years' experience with Arco, while the youngest was just a year out of graduate school. Townsend recruited a combustion specialist from the auto industry and an expert in atmospheric modeling from Caltech. Several team members were avid environmentalists. Says Townsend: ''We felt like we were on a mission from God.'' R&D breakthroughs don't usually happen when an engineer is alone in his office or lab, hunched over a bubbling beaker or a computer terminal. Most brilliant ideas pop up during furious brainstorming sessions. The members of Arco's clean fuels group argued heatedly at weekly meetings and during spontaneous conferences in the hallways. ''They got amazingly combative with each other at times,'' Townsend recalls. ''It definitely wasn't a tea-and- cookies atmosphere.'' In between contentious meetings, team members analyzed a medley of gasoline blends and wrote a computer program that calculated the cost of refining them. In just two years Townsend's team developed three low-emission fuels, and Arco is already selling two of them. For his efforts, Townsend got a big promotion. He's now director of a joint electricity cogeneration venture between Arco and Mission Energy Co.

Engineers are uncommunicative nerds, more comfortable around computers than people, right? Well, it's hard to get a word in edgewise when you're chatting with the chemical engineers at Du Pont, for example. In Wilmington, Delaware, several dozen of them are racing to meet a late-1995 deadline set by President Bush for developing substitutes for chlorofluorocarbons (CFCs), the chemicals thought to be largely responsible for depleting the earth's ozone layer. Du Pont, the country's largest CFC producer, has already concocted less damaging substitutes that can be used as coolants in refrigerators and air conditioners. Du Pont's engineers will work with counterparts from Whirlpool, GE, Carrier, and other manufacturers to redesign their products to accommodate the new compounds. A new buzzword describes this coordinated approach to R&D: ''concurrent'' engineering, which puts a premium on engineers who can communicate well. Says Helen Connon, 41, a chemist who co-heads Du Pont's refrigerants lab: ''It's not enough to be good technically. You have to be articulate to be in my group.'' Her engineering staff includes an Oklahoman who likes to write limericks and an ebullient Englishman on loan from Du Pont's Brazilian subsidiary. When they're not testing the properties of refrigerants and foam- blowing agents, they swap verses by E-mail. Connon is soft-spoken, matter-of-fact, and extremely bright. After earning her Ph.D. in organic chemistry at Yale, she taught for a few years at Berkeley, but the academic life left her cold. So she jumped to Du Pont and found herself instantly immersed in the intricacies of making nylon. After a decade in various labs, she unexpectedly became head of the refrigerants group. Says she: ''One of the lab directors called me into his office and said, 'You're the new boss.' I guess they thought I had good people skills.'' EXPERIENCED technical people often get pushed into managerial jobs. Sometimes, as in Connon's case, it happens at the behest of the company. And some engineers, especially those with children nearing college age, reluctantly become managers simply because they need more money. Many companies in recent years have created parallel ladders that allow engineers to receive substantial promotions and raises without switching into management. But most of the parallel ladders don't reach high enough. While engineers enjoy relatively good starting salaries -- currently in the $35,000 to $40,000 range -- their salary growth afterward is sluggish. The median salary for an engineer with a master's degree and ten years' experience is $53,400. After 25 years, it's $64,850. Even at the best-paying companies, engineers with no management responsibilities rarely earn more than $100,000 a year. Although companies need managers with technical expertise, some engineers just aren't suited for supervisory roles. The ideal company would give them the freedom to stay on the technical side as long as they wanted. Microsoft, the largest maker of software for personal computers, comes close. From its founding 17 years ago, Microsoft has been a company of programmers led by a programmer -- Gates, who got his start writing software in the Basic computer language. At Microsoft, programmers are the company's stars. They are a colorful bunch. David Weise's office at Microsoft headquarters in Redmond, Washington, looks like a graveyard for deceased PCs. Circuitboards and other computer innards are scattered all over the place. In one chair sits a cardboard skeleton that Weise, 34, constructed from a kit in his spare time. Weise talks very quickly; the speed of his soliloquies can be measured in megahertz. He wears a T-shirt with Maxwell's four laws of electromagnetics written on it in abstruse mathematical symbols. Weise paid his way to a Princeton Ph.D. in biophysics by playing blackjack. He then helped form a high-tech startup in the Bay Area. Microsoft bought the fledgling company in 1986; Weise has since been working on Windows, Microsoft's best-selling operating system for PCs. His specialty is graphics -- for example, making the colors and text look as good as possible. Says Weise: ''Writing software is a craft like woodworking or masonry. It takes good judgment to do it right.'' He spent most of last winter debugging the Windows 3.1 release, a process akin to detective work. The toughest bugs to fix, according to Weise, are those overlooked in earlier releases of the software. Says Weise raffishly: ''It's like sex. You make a mistake once, and then you have to live with it for years.'' Microsoft's culture encourages flexibility. Programmers can become managers if they want to, or they can become ''technical leads'' who guide a project but don't evaluate or supervise their teammates. Konzen, 30, who started working for Microsoft right out of high school 12 years ago, decided the management route was not for him. Says he: ''I was doing a lot of managing, and it got to be a bummer after a while. I was getting away from the programming, the hands-on stuff that I like. So I moved back to the technical side.'' Konzen didn't have to worry much about the financial consequences: Since Microsoft has a generous stock option plan and the company's share price has grown 20-fold over the past six years, programmers like Weise and Konzen are already well off. Flexibility at Microsoft sometimes verges on chaos. Programmers come to work at all hours, usually in jeans, T-shirts, and sneakers. Konzen keeps a couple of electric guitars in his office. He also likes to race Porsches and build homemade bombs, a talent he picked up in junior high school. He once taught colleagues how to make rocket fuel in an office kitchen. They later used the fuel to blow up a traffic cone in one of the company parking lots at 2 A.M. When security guards showed up with bomb-sniffing dogs, the programmers fled to the company library. When they're not building bombs or racing Porsches, Microsoft's programmers work like hell, driven by a lively competitive spirit. Windows is vying with IBM's OS/2 software for the chance to replace MS-DOS as the standard operating system for PCs. Back in the days when IBM and Microsoft were the closest of allies, many Microsoft programmers worked on OS/2, and that experience gives them extra confidence now. Says Konzen: ''We know how to beat our own code.'' Although the outcome of the battle won't be known for a while, most industry analysts are betting on Microsoft. Many companies that face intense global competition are trying to inject that fighting spirit into their engineers. General Motors has assembled a team of 200 people to develop an electric car called the Impact. GM tried to make an electric car once before, in the early 1980s, without much success. But by taking advantage of recent breakthroughs in battery technology and microelectronics, GM engineers believe they can produce an affordable electric car by the mid-1990s that will go up to 125 miles without a recharge. A typical GM believer -- zealot may be the better word -- is Jon Bereisa, chief engineer for the Impact's propulsion system. Bereisa, 47, is accustomed to working on futuristic vehicles. After getting a master's in electrical engineering at the University of Missouri, his first job was with a NASA- sponsored project to develop a nuclear rocket engine. ''It was just like one of the engines on the Starship Enterprise,'' says Bereisa, a lifelong tinkerer who spent his teenage years disassembling lawn mowers and any other machinery he could get his hands on. He joined GM in 1974 and helped develop the microcontrollers now used in virtually every automobile to limit emissions, regulate the power train, and perform a multitude of other tasks. On the Impact project Bereisa is an idea man; he continually churns out proposals and suggestions. ''I like being busy,'' Bereisa says. ''For me, hell is a place where there's nothing to do.'' Bereisa's counterpart on the manufacturing side is Bill Szkodzinski, manager of vehicle assembly systems for the Impact. Szkodzinski, 37, has been with GM since high school, when he turned down a chance to join a semiprofessional hockey team. He received a bachelor's degree in mechanical engineering from the GM Institute, a college in Flint, Michigan, formerly owned by the automaker, and then worked in several assembly plants, developing ways to build cars more cheaply and efficiently. Manufacturing engineers are a different breed from design engineers; they generally have less education but more experience on the factory floor. Says Szkodzinski: ''I'm not the kind of guy who can sit in an office and produce a drawing. I thrive on meeting people and getting my hands dirty.'' Design engineers in the U.S. have traditionally enjoyed better pay and more prestige than manufacturing engineers. But as American companies get serious about improving product quality, the status of manufacturing engineers is rising. At GM's Impact project, design engineers and manufacturing engineers work in the same offices, usually within ten feet of each other. As soon as the designers begin work on a fender or headlight, the manufacturing experts start figuring out the least expensive way to make it. Result: faster development and less rework. Even when a company is losing badly to its rivals, a top-notch engineering effort can help it leapfrog the competition. Xerox spent most of the Eighties getting trounced in the copier market by Canon and other Japanese rivals. This year Xerox finally struck back with the 5775 digital color copier, which spews out 7 1/2 color copies a minute -- 50% faster than Canon's best machine. The key is a unique paper-moving device called the twin roll transfer loop (TRTL), or Turtle for short. Rick Dastin, 33, a mechanical engineer with a master's degree from Rochester Institute of Technology, led a five-year effort to turn the clumsy Turtle into a speedy hare. Since the Turtle used radically new technology, its development lagged far behind the rest of the machine's. Says Dastin: ''It was a real nightmare at the beginning. The thing was just a piece of junk. We were ready to dump it many times. But the knowledge of its potential advantages kept us going.'' In slow, dogged steps, Dastin's team taught the Turtle how to grip a piece of paper and move it along. They eventually got the Turtle's accuracy in paper placement down to within 50 microns, less than the width of a human hair. Good engineering also involves market research. Before starting the Turtle project, Dastin made several trips to talk to customers who were using Japanese copiers. He learned that customers wanted a machine that could make color copies on heavier paper, for report covers and the like. Incorporating this feature into the 5775 became a top priority for Dastin. Says he: ''It didn't matter that I had no budget for it. I went ahead and did it anyway, because I knew it would help make our machine the best.'' The 5775 is the only color copier on the market that can print on 90-pound index stock, the paper that index cards and document covers are made of. CRAY RESEARCH, the largest American maker of supercomputers, got a late start in massively parallel processing. Massively parallel supercomputers combine the computing power of thousands of microprocessors, enabling them to run faster than conventional supercomputers. Intel, Thinking Machines, and several other companies have already demonstrated massively parallel supercomputers, but Cray didn't decide to make one of its own until last year. Cray has some competitive advantages, though, and one of them is supercomputer designer Mark Birrittella, 37. As a boy, Birrittella made the common mistake of confusing engineers with the people who run locomotives. A big train-lover, he was disappointed when he recognized his error, but he decided to become an engineer anyway. After receiving a master's degree in electrical engineering from the University of Florida, he specialized in designing integrated circuits, first for Motorola and then for Cray. Along with 30 other hardware engineers, he's drawing up plans for Cray's massively parallel machine. The designs are written in Boolean logic, an esoteric mathematical shorthand. Says Birrittella, who has two young children: ''Sometimes when I come home from work, I'll still be thinking about the design. My kids will say, 'Daddy, you're not paying attention.' And the next day it'll take an hour to get back into the mental state that I was in the night before.'' Like Konzen at Microsoft, Birrittella tried his hand at management and found he didn't like it much. Until last winter he was supervising 20 people in the massively parallel design group. In that role he was responsible for assessing the staff's work, arbitrating disputes, and making sure the project stayed on schedule. But it left Birrittella little time for hardware designing, the task . he enjoyed most. ''I was constantly torn,'' he says. ''I missed being on the technical side. I missed just closing my door and doing some real design work.'' After much angst and soul-searching, Birrittella left the supervisory role and is happily writing Boolean logic again. That's what's important: Keeping your engineers and programmers happy. Because when they are happy, they tend to churn out inventions and breakthroughs and product improvements and all sorts of other nice things. Just remember that what makes them happy may not be what appeals to you or other managers. Companies that accept this fact -- and act on it -- are the ones that manage engineers best.