Stream Analyze, headquartered in Sweden, describes itself as a provider of ‘AI everywhere’ technology. The company’s story is a fascinating one – and it is using edge computing to deliver on its promise.
Co-founders Tore Risch, Erik Zeitler and Jan Nilsson joined forces in 2015 with the mission of turning Risch and Zeitler’s research around real-time data analysis into a product.
The company takes up the story. “Initially, the computing was hosted in the cloud, in line with the major current AI and IoT trend, but the founders soon started to recognise the unique advantage of the small footprint of the software, since it would allow the technology to be installed on any device.
“This is where the edge-based strategy of Stream Analyze started to take shape.”
The company’s offering, put simply, is to install its entirely agnostic software on ‘essentially any device out there’, be it a car, a chainsaw, or a manufacturing device, to make it smart, initiate a dialogue, and interact and analyse in real-time. The only proviso is that the device needs a processor, some memory and its own connection port, be it Wi-Fi or 4/5G, to which the software piggybacks.
Daniel Spahr, chief operating officer at Stream Analyze, says that the company refers to the cloud as ‘stale data’, compared with processing at the edge. “What we do is analysis of streamed data as the data happens, digesting it as close to the data source as possible, only pushing up what is necessary into the cloud,” Spahr tells Edge Computing News. “You’re saving time and you’re saving money, by doing this.
“It’s sort of ‘Excel on the edge’,” Spahr adds. “You have your Excel spreadsheet, you open it up, and then you build your own models, whatever you like. You don’t have to be a super programmer to work with it.”
Spahr was a college friend of CEO Nilsson, and knowing he wanted to move to a company building something from scratch, the chief executive recommended Stream Analyze.
His career journey to date has run the gamut of telco, cloud and IoT. A previous role was CIO at a global pest control company, Anticimex Group, where the goal was multinational growth through digitisation – “leveraging business through IoT, deploying digital traps” as Spahr put it.
The move to edge was therefore a logical one – but this is no smokescreen or marketing gimmick. “The definition of edge can vary,” Spahr notes. “To some it’s just ‘not cloud’ – but the tier that we operate is what we sometimes call true edge or micro edge. You’re as close to the sensors as you can actually get.”
A current customer example can be found through a global forklift manufacturer. The company makes up to 20,000 service visits per day on the forklifts, and approximately 80% of the cost of upkeep is accrued by the manufacturer itself. “It’s very reactive”, Spahr notes.
Through Stream Analyze, the manufacturer is now able to make two adjustments. The first, which is obvious enough, is to add predictiveness, for instance if a bad noise or temperature increase is spotted in the sensors. The second is to add greater remote capability. If a forklift breaks down, a technician can drill down onto the vehicle and see what is broken – or in some cases, even fix it remotely.
When the company’s journey to edge began, Spahr notes that education was needed, even among early-adopting customers. But the explosion of IoT, and the serious amounts of data being aggregated, has had its effect.
One automotive organisation Stream Analyze works with is collecting more data than even their wildest projections suggested. An executive, Spahr notes, ‘said the amount of data [they] expected to capture was many multitudes higher than what [they] even calculated.’ “Pushing all that data up into a cloud is just not feasible,” he adds. “They need to do that analysis on the car directly.”
This has been known for some time. Speaking to Jonathan Reichental, then CIO of the city of Palo Alto, in 2016, he noted how organisations across smart cities – automotive being a prime example – were ‘going to have the typical big data problem.’ “We’re gathering more than we need, and it’s in abundance, so what do we do with it?” he said.
Ultimately, Spahr sees Tesla as a cornerstone example of how the industry is transforming. “Tesla has driven the automotive industry into a huge change, because what Tesla’s saying is a car should be electrical, it should be powerful – it should be software on wheels,” says Spahr. “This is disrupting everybody, because they’re now moving away from mechanical engineering to software engineering.”
This use case was so alluring that Spahr spoke about it at the inaugural Edge Computing Expo event in February. His session explored how Tesla had set a global standard for continous product development – and how by utilising edge AI technology, companies could do it as well, or better, than them.
As Stream Analyze will be participating in the next leg of the series, the Edge Computing Expo Europe on June 16-17, this is the message the company is looking to get across. “We can help car companies do what Tesla is doing in a very accelerated way,” explains Spahr.
“You can’t ignore this any more,” he adds. “It needs to happen, and it’s happening inside your core products. Therefore it can’t just be something that the software department is determining, or engineering departments determine. It’s part of something that the management has to embrace.”
Want to learn more about topics like this from thought leaders in the space? Find out more about the Edge Computing Expo, a brand new, innovative event and conference exploring the edge computing ecosystem.