On Friday, President Obama celebrated Darpa, the Defense Department’s premiere research arm, as a hothouse of “cutting-edge ideas to create new jobs, spark new breakthroughs, [and] reinvigorate American manufacturing.” With agency director Regina Dugan at his side, Obama inspected a new armored vehicle — crowdsourced under Darpa’s auspices, and built and designed in just 14 weeks. “Not only could this change the way the government uses your tax dollars,” Obama beamed, “it may save some lives in places like Afghanistan for our soldiers.”

Lost in the celebration was that the Senate Armed Service Committee, just days before, recommended drastically cutting back Darpa’s budget — and singled out the crowdsourced vehicle project as a problem.

The $150 million proposed cut — more than five percent of Darpa’s $2.98 billion budget — not only pits a powerful Congressional panel against a White House which has made science and technology a priority. It also undermines a key component of Dugan’s management of the agency.

For years, Darpa clashed with Congress (and with its Pentagon superiors) about how the agency spent its money. Former agency chief Tony Tether would personally review the hundreds of research efforts in Darpa’s portfolio. And if those projects weren’t progressing according to plan, Tether would withhold their funds until they caught up. That drove Capitol Hill nuts, because it left tens of millions of dollars on the table. What Tether called a simple matter of accountability, Congress called micromanagement, “unobligated funds,” and “poor execution.” Every year, it seemed, one Hill panel or another recommended big budget cuts.

That was all supposed to change when Regina Dugan took over at Darpa. A former entrepreneur, she made it a priority to install a more business-like operating environment at the agency. No more micromanagement. Day-to-day operation of the agency was left to Dugan’s deputy, Kaigham “Ken” Gabriel. Decisions about individual research programs were pushed down to the agency’s six office directors. And those programs should be largely allowed to run their course, without having their cash flow threatened every six-to-eighteen months.

Last year, at a reunion of Darpa program managers — an event usually reserved for celebrating the agency’s latest bleeding-edge innovations — Dugan and Gabriel instead trumpeted the fact that all of the agency’s funds had been “fully obligated.” It struck several Darpa alumni as a rather tepid achievement for the agency that gave rise to stealth technology and the internet.

Nor does it appear to have mollified Darpa’s paymasters at the Senate Armed Services Committee. Darpa has never been quite able to spend all of its allotted cash, the committee notes in its report on next year’s Pentagon budget, obtained by InsideDefense.com. Why should this year be any different?

“While Darpa’s fiscal performance has notably improved, the committee is still concerned about the timeliness of sustained funding execution,” the report observes. “The committee recommends a reduction of $150.0 million from Darpa’s overall budget to reflect continuing concerns about timely and effective execution of funds by the agency, as well as concerns about specific programs.”

Those specific programs include some of Darpa’s more outlandish efforts, like the ones to assemble a flying car, recruit animals into an online network of “unconventional warfighters,” and build portable nuclear reactors for battlefield outposts. The panel, in a feat of understatement, notes that there are “a broad spectrum of policy and regulatory issues associated with deploying a small nuclear power source to a FOB [forward operating base], or other remote location.” The committee also invites Darpa to reconsider “safety and security issues, fuel cycle and other sustainment issues, as well as issues regarding public relations and strategic communications that would have to be addressed when deploying such a system to a host nation.”

It’s easy to understand the reservations about such seemingly marginal projects, especially when Washington is in a budget-cutting mood. But the committee also took aim at the program heralded by President Obama and placed at the center of Dugan’s strategy for Darpa.

The crowdsourced “Fast, Adaptable, Next-Generation Ground Combat Vehicle” project may be a good way to infuse the staid defense industry with new ideas, the Senate report argues. But the crowd isn’t necessarily an authority on what’s really needed to keep troops alive in places like Afghanistan. The committee is concerned that these so-called “force protection and related armor technologies” are “not adequately address[ed]” in the FANG effort.

FANG is one of the early results from what is supposed to be a massive production line of Darpa programs to reinvigorate American manufacturing. The agency is planning to spend a billion dollars over the next five years on efforts like FANG. That’s a huge number for an agency that spends three billion a year.

Of course, Darpa’s mission isn’t to jump start sectors of the U.S. economy. Not directly, at least. It’s to continue American military’s technological advantage over its rival. Those two jobs are one in the same, Dugan told Congress last year.

“Americans today consume more goods manufactured overseas than ever before, and yet they are less likely to be employed in manufacturing than at any time in the last 100 years,” Dugan said. “This has potentially significant implications for defense, because to protect, we must produce.” But in order to produce, Dugan first has to convince Congress that her manufacturing program — and her agency — is really on track.