As it celebrates its 75th anniversary, the Skunk Works is in the middle of a huge and unexplained growth spurt.

The Palmdale, California-based unit of Lockheed Martin focused on the most advanced and secretive projects employed about 3,000 workers last year. It has about 3,500 employees now. And that’s about halfway to a near-term goal of about 4,000. On-boarding sessions for new employees used to be held every other week, but are now scheduled twice a week — and they’re full.

“The number is moving so quickly we hesitate to give a number,” says Craig Johnson, the Skunk Works’ director of Business Strategy and Development, during a meeting with reporters in Palmdale.

The reason or reasons behind the hiring spree — representing a 33% jump over the 2017 payroll on top of retirements and departures — is not totally clear. It may be an indicator of a recent major programme launch.

So far, the Skunk Works isn’t saying. The unit operates as a division of the Aeronautics business sector within Lockheed. Its annual economic activity of more than $1 billion this year — roughly equalling the Skunk Works’ heyday at the height of F-117 production — is about the same size as the F-22 modernisation programme within Lockheed’s Aeronautics sector, Johnson says.

He showed a briefing chart that demonstrated the diverse range of the Skunk Works' activity, but no obvious signs of a major new production programme.

It included images of the compact fusion reactor, an effort Lockheed acknowledged five years ago to develop a breakthrough nuclear powerplant. But Lockheed officials have said ongoing tests on a series of subscale prototype reactors won’t produce data needed for a go-ahead decision until later this year or next year. Other images showed a next generation fighter, which is in the early stages of the Department of Defense’s acquisition planning process.

Another image presented the ducted-fan powered ARES unmanned vehicle, which Lockheed hopes to fly at a test range in Arizona by the end of the month. Lockheed plans to submit the ARES as a candidate for the US Marine Corps’ MUX programme, but that is several years away from a contract award. Still another image showed a concept for a next-generation unmanned air system, which represents Lockheed’s vision to replace the General Atomics Aeronautical Systems MQ-9 with a more survivable platform.

Another image showed Lockheed’s vision for a future hypersonics vehicle. Russia and China have acknowledged major advances in testing hypersonic weapons over the last two years, leading some to question the US government’s ability to respond after decades of apparent hypersonic neglect. But Johnson hints that US hypersonics development is farther along that some have reported.

“I don’t believe we are as far behind as some of the press may lead you to believe,” says Johnson. “Lockheed Martin has a very substantial programme going on in terms of hypersonics.”

On 18 April, the US Air Force awarded Lockheed a contract worth up to $928 million to develop the air-launched Hypersonic Conventional Strike Weapon (HCSW, which is pronounced “Hacksaw”). The contract was warded to Lockheed’s Denver-based Space Systems division, but Johnson says that Skunk Works is heavily involved in the project.

“Skunk Works led the effort,” Johnson says. “There is tremendous collaboration with the Space Systems” sector.

Meanwhile, Lockheed also is turning its attention to solving one of the US military’s toughest problems. A revolution in machine-learning software algorithms in the commercial market has bypassed the most advanced combat aircraft because their processors use archaic architectures that take years to accept new upgrades. The solution, according to the Skunk Works approach, seems simple: accept the US military’s new open mission standards as the foundation for an operating system architecture, then partner with start-up companies with innovative software applications to rapidly roll-out new capabilities.

During a tour of the Skunk Works Integration Facility and Test Laboratory on 14 June, John Clark, vice-president of intelligence, surveillance and reconnaissance, demonstrated how such an architecture could work. Using a surrogate simulator for a fighter cockpit, the “pilot” defined an area on a map, then pressed an “automatic target recognition” button on the cockpit display. The command cued a new, state-of-the-art processor to query the fighter’s sensor data and automatically recognise several targets within that area. The same architecture also has applications that can allow different aircraft types to share data with each other.

The Skunk Works architecture is making rapid progress. The organisation has completed a five-year series of demonstrations with enigmatic titles, such as Project Missouri, Project Iguana and Project Hunter. The most recent event, the System of Systems Integration Test, and Experimentation (SOSITE), was funded by the Defense Advanced Research Projects Agency.

It’s no longer just an experiment. The USAF also has approved a plan to integrate Lockheed’s architecture into the F-22 fleet as part of the Tactical Mandates (TACMAN) and Tactical Link (TACLINK) upgrade programmes. For the first time, the USAF’s front-line fighter will be able to use machine-learning algorithms to process the volumes of data collected by the F-22’s powerful sensors. A version of the same architecture also is being integrated with the F-35 as part of the Technical Refresh 3 programme.

Clark offers the OMS as an example of the Skunk Works’ unique methods in operation. By relying on available technology and a series of flight demonstrations, the Skunk Works has the tools to push new capabilities into combat faster than others, he says.

It’s a culture and a process that must be protected as the organisation grows by one-third over less than two years, says Rob Weiss, the 10th executive vice-president and general manager of the Skunk Works. Weiss is retiring from Lockheed in December, but is handed over leadership of the Skunk Works to Jeff Babione on 14 June.

“You have to protect this culture here at Skunk Works,” Weiss says. “You have to take the goodness the larger organisations have to offer, but limit it so you can preserve the ability to be quick and be affordable.”