IEEE Spectrum Geek Life Channel
-
From Spacecraft to Sensor Fusion
It's easy toset up multiple video feeds in all kinds of locations. Merging those feeds together, combining them with other information, and putting them into context is a tougher challenge, especially when your customers include first responders and intelligence agencies. But that's the job that Cubic Corp. hired Iverson Bell III to do in late 2020, when they chose him “to lead the team responsible for transforming the company's Unified Video project into a more full-featured video/communications platform, including AI/machine learning and support for distributed sensors and more data types," says Bell.
Bell had previously been at Northrop Grumman, doing spacecraft design and testing, and had researched electrodynamic tethers, including for use with CubeSats, as part of his postdoc work with Brian Gilchrest at the University of Michigan, where Bell earned his master's and Ph.D.
Moving to his role at Cubic's Hanover, Md., location might seem like a big jump between two very different specialties, but Bell was able to leverage common skills and other experiences in dealing with complex data handling and processing. For example, Bell created sophisticated data-analysis models during an internship at the Johns Hopkins University Applied Physics Laboratory; performed software testing along with electrical-system integration and testing for the James Webb Space Telescope program at Northrop Grumman; and has been an agile lead manager on other software development.
The path to his current job and project, according to Iverson, reflects a mix of exposure, opportunities, and risk-taking—both by him and by people in charge of him—plus, of course, a lot of hard work.
“My undergraduate engineering focus at Howard [University, in Washington, D.C.,] was applied electromagnetics, like antennas and waveguides—so there was lots of math involved. And I was also interested in signal processing, like how antennas convert data into a digital signal. Then, in grad school at the University of Michigan, my concentration was electromagnetics, and I got a lot of exposure to remote sensors, radars, and other cool stuff in more depth.
“I chose good topics by pure luck," says Bell.
Bell attributes his interest in science and engineering to early exposure through books, other media, and family. “I read books like Marshall Brain's How Stuff Works series, and watched the Discovery Channel," he recalls. “My mother is a pediatrician. And she was a chemistry undergrad, and cooked—and cooking was chemistry. And my older sister has been a role model—she's a civil engineer—and I was exposed to her taking classes. When I got to high school and was liking math and science, she asked me, 'What do you want to major in?'"
“Be ready to take risks and learn as you go," Bell advises. “For example, when I was part of the group conducting electrical tests on the Webb telescope, we were working with the mechanical engineering and test team, which I hadn't been exposed to previously. It was something to learn.
“It helps when folks are willing to take a chance on you—so if you're in a position to employ someone, take that chance on them…. I have a passion for mentorship," says Bell. “I want to help improve education for younger people, expose them to STEM topics.... But my interest is shifting from individual mentoring to wanting to help address the larger policy and curriculum aspects of the problem."
Bell is enjoying the shift from the process of building spacecraft—“where the process takes time and you often get only one chance for things to work"—to delivering a real-time service. It's “very different to see teams working on high-end leading-edge engineering at a very fast pace," he says. “It's like changing the plane's engine while you're flying it."
-
Engineering in Secret
When your job involves working on sensitive information, products and projects, how do you talk about it, not just at work, but also at conferences, when mentoring or recruiting, or at dinner and social events?
“When information about what you are doing is privileged—classified—you can’t talk about it,” says Dennisa Thomas, senior surety systems engineer at Sandia National Laboratories “But you can talk about the general understanding or expertise you have gained from certain systems and materials projects, and how you can transfer that knowledge and those skills to other spaces.”
One definition of surety, according to Thomas, is “a level of confidence that a component or system will operate exactly as intended, both under expected and unexpected circumstances.” This includes not just Sandia’s mandate to keep the United States’ nuclear stockpiles safe, secure, and effective, but now also tackling complex national security problems including homeland security, transportation, energy, and cyber-, chemical and biological defense.
For example, reports Thomas—who has worked on hundreds of components and systems—“I worked with the team that put the first Sandia-designed telemetry transmitter into production. I’m currently working on qualifying two fusing/firing assemblies for production.”
Thomas got into surety “by picking opportunities,” she says. “North Carolina Agricultural and Technical State University (NCA&T), where I got my B.S. in electrical engineering, has a large career fair twice a year.... During my first year, I was offered an internship with the NSA, where I learned about some of the communications systems and work they do for the military.
“And at another career fair, somebody from Sandia spoke with me, telling me about their Masters Fellowship Program, which gives graduating seniors a chance to attend graduate school to achieve their master’s degree in an area of focus that Sandia is interested in. I went to Florida State University for my M.S. in electrical and electronics engineering and then in 2015 came back to Sandia full time.”
Surety appealed to Thomas because “you get to see how the pieces and teams all fit together.” For those interested in the field, even outside government work, “Learn about failure analysis,” says Thomas. “For electrical engineering, math and science is a given. Having a strong foundation in circuit analysis and electronics is important. And statistics is important—if we can’t interpret the data that’s collected, it’s not as helpful.”
“It’s also essential in quality and surety to learn how to be a good team player, [to] develop great team-working skills,” she stresses. “Slow communications of problems will ultimately delay the right solution. Soft skills are important so you can communicate there is a problem in a process that could have been better defined, without finger-pointing.”
Mentoring and recruiting has also been an important part of Thomas’s activities. “I have become one of the recruiting leads for Sandia’s Securing Top Academic Research & Talent for Historically Black Colleges and Universities (START HBCU) initiative,” says Thomas.
Before moving to Kansas City, where Sandia collaborates with the National Security Campus, “I volunteered at the First Lego League challenge at NCA&T. Where I am now, we go to schools for ‘Introduce A Girl To Engineering Day,’ and I work with my sorority to introduce kids to STEM fields—for example, I taught a class on how to build a penny battery.”
“Nothing makes me more excited than seeing children develop their craft and introducing girls to engineering. Having people who look like them is invaluable to developing the next generation of engineers and applied scientists.”
This article appears in the July 2021 print issue as “Engineering Secrets.”
-
This Tech Entrepreneur Scored With Home Media Streaming; Now He’s Miniaturizing the Chocolate Factory
With a rattle not unlike the sound of coffee beans being dumped into a grinder, Nate Saal pours a scoop of cocoa nibs into the top of the latest version of his chocolate making appliance. It takes up less than a square foot on the countertop, and fits easily under most cabinets.
It wasn’t easy to shrink a chocolate factory into a small gadget. It’s been more than five years since Saal made his first attempt, now sitting on the floor in front of the counter and began turning his quest into a company, now named CocoTerra and with several million dollars of angel investment behind it. Four more prototypes followed.
Starting a company wasn’t new to Saal. After graduating with a degree in molecular biophysics and biochemistry from Yale, he became fascinating with the possibilities of the web, first joining Smart Valley, a non-profit aiming to increase access to Internet technology in 1994, then starting his first web-technology, CatchUP, an automated update system, two years later. He sold the company to CNET in 1999.
Next came GlooLabs, and a media streaming platform. Cisco purchased that company in 2007, adding the technology to a Linksys network storage device tagged the Media Hub. Then came Tactus Technology, a startup developing a touch-screen interface that used microfluidics to create a morphing keyboard with temporary raised buttons. In 2013, with funding dwindling and smart phone manufacturers committed to glass screens, he left that company.
What to do next?
The idea for a chocolate machine came during a visit from his brother-in-law. “He’s in the coffee supply business,” Saal said, “so I thought it would be nice to take him to a chocolate tasting; chocolate and coffee make a pretty good pair.”
At the tasting, Saal recalls, they discovered similarities between the two industries: where the product is grown, that it is hand harvested, that they are both roasted. What was missing was the ability to make chocolate at home, without any expertise.
Saal wanted a device to let him do that, if it existed.
“People love to make what they love to eat,” he says. “So you have ice cream makers and coffee makers and popcorn makers and pasta makers and bread machines, why not chocolate?”
If he couldn’t find one, he thought, maybe he could build one. He decided that the starting ingredient would be nibs, cocoa beans that have been de-shelled and oven roasted, taking a page from gourmet coffee, which generally starts with roasted beans.
TEKLA PERRY
First to figure out: how to grind the nibs. Saal decided to use what is called a ball mill, that is, stainless steel balls tumbled by a motor. He chose this chaotic-motion grinding process, he says, for its efficiency, important for a process intended to operate quickly in a small device. Grinding the nibs releases the fat in the beans, Saal explained, which must be kept warm so that the chocolate doesn’t solidify before the rest of the ingredients can be added.
Next came chocolate making’s trickiest step: controlling the crystal formation through tempering. Cocoa butter can crystalize into six different structures, only one of which is desirable, so the temperature must be carefully controlled. For this, Saal settled on thermoelectric coolers, solid-state devices that get extremely cold on one side and hot on the other when a current runs through them.
Finally, he needed the machine to move the liquid chocolate into a mold. Commercial chocolate factories generally use pumps and dispensers for this. Saal, a beekeeper since high school, developed a centrifugal molding process, similar to that used to extract honey from honeycombs. This allows the machine to use the same motor to extract as it uses to grind, an advantage that overcomes the possible negative of a slight curve in the final bars.
COCOTERRA
That’s the basic machine design. The whole thing can be controlled by buttons on the machine, or, more intuitively, by an app that includes preloaded recipes, taste quizzes to help users develop their own recipes, alerts indicating when ingredients like sugar and milk powder need to be added (the only part of the process that isn’t automated), and status updates. This automation is essential to those new to chocolate making, Saal said, but he also left the option for more experienced makers to control many of the parameters directly. (Ultimately, the app will include a marketplace for cocoa nibs and other ingredients and a way of sharing recipes with other users. )
Saal says CocoTerra will ship products early in 2022; the company is now taking preorders at $799 and intends the initial sticker price to be $1199. He expects the price to come down with time to compete with other countertop appliances (a typical KitchenAid mixer, for example, retails at around $400. Cuisinart’s Complete Chef, that cooks as well as chops, sells for around $600.)
While the hardware design has been basically complete since last year, ramping up to manufacturing hasn’t been quick.
“This is something that hasn't been made before,” he says. “It’s a new category; there are no existing sub-assemblies we can leverage. We have to make it all, and it’s a complex mechanical system.” Unfortunately, the company missed a perfect marketing opportunity—the wave of home-made everything that hit with the stay-at-home orders of the pandemic.
Credit
Saal and I had been sitting outside for most of our conversation, following Covid protocols. As we finished our conversation, he checked his phone; the chocolate was just about ready. We stepped back into the house to wait for the machine to finish its final steps, spinning the chocolate into the mold, and then continuing to spin it as it cooled.
And it was time to taste. I’m a dark chocolate fan, so was happy that Saal had selected a 70 percent cacao recipe, using nibs from Madagascar, for the demo. The result—creamy, chocolatey, slightly fruity (that’s Madagascar’s signature)—and good. Really good. Insanely good.
“That’s because it’s fresh,” Saal explained. “Everything that people buy in the store has been there for weeks, at least. I don’t want to hype it, but the experience of fresh chocolate—it’s amazing.”
What we didn’t eat, Saal packed up for me to take home. Instead, I stopped to do a reality check with friends who are serious foodies, and chocolate is one of the foods they are serious about. Their verdict? “OMG, just, OMG.”
Does Saal have another winning startup? He will start finding out early next year, when consumers can decide whether this gadget deserves a place on their countertops.
-
Where Are the AI Jobs? Look to a Farm or a Forest
Where are the AI jobs in the U.S.? The largest share, of course, is in the IT industry, with professional, scientific, and tech services coming in second place. But coming on strong are fields in which you might not expect to see large number of AI and machine learning professionals—agriculture, forestry, fishing, and hunting, according to an analysis of 2020 job postings.
For AI is increasingly being applied to forest conservation and management. Meanwhile, farm equipment maker John Deere put big and early bets on machine learning, and other ag-related businesses large and small are using AI for soil analysis, monitoring crop health, planning planting cycles, and a host of other purposes.
Graph:Tekla Perry/IEEE Spectrum
The analysis, using data gathered by labor market research firm Burning Glass Technologies from some 45,000 online job sites, was reported in the 2021 Artificial Intelligence Index, an extensive roundup of AI trends published by Stanford University’s Institute for Human-Centric Artificial Intelligence.
The analysis also found a several other industries outside the tech mainstream who are also hot on the hunt for AI professionals, including public administration, mining, real estate, and food services:
AI INDEX REPORT 2021
Burning Glass also looked at changes in demand for specific AI-related skills, grouping them under seven general areas: machine learning (including recommender systems and classification algorithms) general artificial intelligence (things like expert systems and IBM’s Watson), neural networks, natural language processing, robots, visual image recognition, and autonomous driving. Machine learning started out strong and grew stronger, and demand still remains far ahead of the rest of the pack, in spite of a recent dip.
AI INDEX REPORT 2021
-
The Endless Frontier Act Could Drastically Change NSF
The United States Congress is considering new legislation that could change the structure of the National Science Foundation (NSF), the agency responsible for funding much of the U.S.’s research. This could cause a major shift in how NSF allocates its research funding—and it’s a shift that could impact tech research.
One chamber of Congress, the Senate, is currently the debate floor for the what was previously known as the Endless Frontier Act, recently rebranded as the U.S. Innovation and Competition Act. Although not all of its proposed funding will go to research, the Senate bill is worth on the order of US $100 billion—several times larger than NSF’s current budget.
Importantly, that money isn’t going to defense-related spending. According to Russell T. Harrison, IEEE’s Director of Government Relations, “It’s targeted in parts of the federal budget that are traditionally not especially big,” including agencies such as the National Institute of Standards and Technology (NIST), the Department of Energy's Office of Science, and the NSF.
The NSF would play host to the bill’s centerpiece: a new “Technology Directorate” with its own independent funding stream. It would get the lion’s share of Endless Frontier’s funding, and that funding would be, for now, mandated to go into research for certain technologies.
In fact, it would be limited to research in ten delineated, if broad, research areas: high-performance computing, quantum computing, disaster mitigation, biotechnology, energy technology, semiconductors, robotics and automation, advanced communication (such as 5G), cybersecurity, and, in Harrison’s words, “anything else that’s necessary to do the other nine things.”
There is some flexibility: It’s possible, in time, to change one area out for another not on the list. And NSF would still be able to back broader research as it does now, deciding which fields get what funds. But these limitations would mark a drastic shift in how NSF operates. It “is Congress’s way of telling NSF, ‘You need to prioritize these ten areas of research,’” says Harrison.
It’s not coincidence that those engineering- and technology-heavy fields are some of the same fields that consistently make headlines. Indeed, in the Senate, lawmakers are selling the bill as a measure of bolstering U.S. research against the specter of China.
But outside the scope of research, some lawmakers who otherwise support science funding have spoken out that it might add more fuel to the fire of Sinophobic racism. Others are concerned that the Act’s funding targets for NSF are simply unrealistic.
Partly for those reasons, over in the other chamber, the House of Representatives, some lawmakers are spearheading another proposal, branded the NSF for the Future Act. It’s more modest in scope, primarily refreshing and renewing NSF’s funding.
Of course that would still give NSF a boost, also ensuring that NSF’s yearly allocation increases to account for inflation. And Endless Frontier notwithstanding, this “is, really, the first comprehensive reauthorization of NSF in more than ten years.” according to Mark Elsesser, Director of Government Affairs for the American Physical Society (APS).
And the NSF for the Future Act creates a new directorate within NSF, called “Directorate for Science and Engineering Solutions.” But unlike the Endless Frontier Act, this directorate wouldn’t necessarily have a technology focus, and wouldn’t have a mandate for its funding to go into particular fields.
This bill would require NSF’s director to pinpoint a few “focus areas.” In the process, the director would need to take into account “societal challenges” such as climate change, socioeconomic inequality—and, still picking at Endless Frontier’s strings, U.S. global competitiveness and national security.
The APS hasn’t taken a public stance on the Endless Frontier Act, but they have endorsed the NSF for the Future Act. “It’s really, in our view, a balanced approach,” says Elsesser. “You’re establishing something new...but at the same time recognizing what some consider the bread and butter of NSF: that fundamental, curiosity-driven research.”
But the rebranded Endless Frontier Act has also been introduced in the House of Representatives. So the NSF for the Future Act will now have to compete with the Senate bill to determine which bill—or more likely, which combination of bills—the House favors.
“It is probably going to be the Endless Frontier Act with some of [the NSA for the Future Act] added to it,” predicts Harrison, “but there’s still a fight over it.”
Any bill that becomes law will have to pass both the Senate and the House. And with NSF’s broad support, this is one of the few bills that is expected to easily pass both chambers in some form. So, in the process, lawmakers might therefore try to attach all manner of bits and bobs to the bill, even those that have little to do with science.
Therefore, while advocates push lawmakers to include all kinds of reforms—attempting to help such things as diversifying STEM to addressing the helium shortage that’s plagued low-temperature research in recent years—it’s too soon to know for certain what exactly the law will mean for individual researchers.