Year: 2018

Mode, a San Francisco-based startup, came out of stealth today with a new approach to software-defined wide area networks they call software-defined core network (SD-CORE), which they say will dramatically reduce the cost of running these networks over traditional methods.

The company also announced a total of $24 million in funding led by GV and NEA to build on that vision. That vision, according to CEO Paul Hawes, involves spinning up private networks very quickly at a much lower price point than traditional networking typically offered by telcos for a high fee.

“Traditional hardware-defined private networking solutions like MPLS guarantee reliability, but are inelastic, hard to manage and costly. Mode Core was built to enhance SD-WAN, leveraging our breakthrough in routing efficiency to deliver the performance and reliability of networks like MPLS, but with the flexibility, elasticity and affordability of a cloud service,” Hawes explained in a statement.

Some use cases that could benefit from this approach include  interactive streaming, multiplayer gaming, real-time machine learning and remote command and control, according to the company.

The company was formed after a research breakthrough by a couple of researchers at Cornell, Kevin Tang and Nithin Michael. They figured out how to characterize network traffic in mathematical terms, which up to that point was thought to be impossible. “Michael came up with the first math-based description of how a packet-switched network works,” Hawes explained.

This allowed him to build a software-defined, automated way to route traffic on each node on the network. “It doesn’t need any intervention from anybody to tell it how to route packets,” he said. Once he had that figured out, it removed the need for more complex and expensive solutions.

This caught the attention not just of networking theorists, but of investors who saw tremendous business potential in their approach. “A number of VCs familiar with networking problems approached them [and encouraged them] to productize it” he said. NEA was the lead investor on the $8.3 million A round and they also got a grant from the National Science Foundation. More recently they got a $16 million Series B for a total of $24.3 million to date.

To make this all work because they aren’t a telco, they built Mode Core and partnered with Ericsson UDN and 100 other partners to provide that networking power that they lack as a startup. You could think of it as a cloud service for wide area networking, allowing companies access to this kind of advanced networking for a price closer to business internet than private WANs.

The robots are here and one company, Starship Technologies, has raised $25 million to bring even more to the mainstream. This latest round of funding includes a follow-on investment from Matrix Partners and Morpheus Ventures. New investors include Airbnb co-founder Nathan Blecharczyk, Skype founding engineer Jaan Tallinn and others.

These autonomous robots can carry items, like groceries or packages, within a two-mile radius. The plan with the funding is to deploy Starship robots in neighborhoods, corporate and university campuses in both the U.S. and Europe.

Starship has also brought on former Airbnb business development lead Lex Bayer as chief executive officer.

“We are at the point where we are ready to start deploying our network of robots at scale,” Starship co-founder Janus Friis said in a statement. “This additional funding, and Lex’s appointment, will allow us to bring our services to market. Lex joins us with a proven track record of growing businesses that change the way we live for the better.”

But Starship is by no means the only company operating in this space. There’s Boxbot, a startup that recently received a permit to test autonomous vehicles with a safety driver, and Nuro.

Starship has previously partnered with on-demand food delivery companies like DoorDash and Postmates to test out its robot delivery service. Last January, Starship partnered with the aforementioned companies for a pilot program in Redwood City, Calif. and Washington, D.C. To date, Starship robots have traveled more than 100,000 miles in 20 countries, across 100 cities.

Prior to this round, Starship raised $17.2 million in a seed round led by Mercedes-Benz Vans with participation from Shasta Ventures, Matrix Partners, ZX Venturers, Morpheus Ventures and others.

After much negotiation with the Trump Administration, Secretary of Commerce Wilbur Ross confirmed this morning that ZTE, the Chinese telecommunications giant, has agreed to a $1 billion fine. That penalty was assessed following an investigation showing that ZTE had violated U.S. sanctions and had provided technology to Iran and North Korea. Ross made the comments on CNBC’s morning show Squawk Box.

This was an incredible scare for ZTE. The Trump administration had previously announced that it was banning U.S. companies from selling ZTE components for seven years, components critical to its entire product line. That decision was expected to kill the company, which employees tens of thousands of Chinese workers and is worth billions of dollars in market cap.

As trade tensions have mounted, ZTE has become something of a sore point in those talks. The U.S. position has been that the fine is a law enforcement action, which should not be included in the discussions around trade. However, some sources who corresponded with me say that the Chinese government has purposely held up the acquisition of NXP Semiconductors by Qualcomm while it negotiates a more positive outcome for ZTE.

Regardless of those talks, the Trump administration is facing a deeply hostile reaction from Congress, where there is bipartisan opposition to the deal. Senate minority leader Chuck Schumer (D-NY) has said publicly that “Both parties in Congress should come together to stop this deal in its tracks.” Senator Marco Rubio (R-FL) has floated a proposal for Congress to block this sort of deal.

This is a developing story

While Uber is growing its business with a net loss on its balance sheet, a smaller rival has confirmed a round of funding, and projects that it will be profitable by Q1 of next year. Gett, the transportation-on-demand startup that competes with the likes of Uber but also traditional taxi cabs, has raised $80 million led by existing investor VW with participation from other previous investors.

Post-funding, Gett is now valued at around $1.4 billion, CEO and founder Dave Waiser told TechCrunch in an interview. This is the first time the company has officially disclosed its value.

It may give Gett a claim to being a “unicorn,” but it is still a pale number when you compare Gett to Uber, which is doing a secondary round right now at a $62 billion valuation. but Gett is playing with another strong card in its hand: Waiser said the funding will be enough to get Gett — which operates in 120 cities globally — to profitability in all of its markets by Q1 of next year, and possibly earlier.

The news comes on the heels of a report that Gett is looking to raise $350 million — a figure that Waiser dismissed, but did not deny outright when I asked about Gett’s plans beyond profitability and this current investment:

“For now, we’re laser focused on this important [profitability] milestone, and the funds we’ve raised to execute on that strategy,” he said, “but when you consider competitive markets like New York and London, we will start thinking about what our next milestone should be, after we make money.” The company has now raised just under $700 million with other investors including Access Industries, Baring Vostok and MCI (Russia’s Sberbank and Kreos have previously provided debt).

In the 120 cities where Gett is now active, Waiser said that New York, London and Moscow are its biggest markets, with half of the company’s volume coming from New York and London alone, on a revenue basis.

Gett, he said, is currently at a $1 billion/year rate, meaning collectively those two cities account for $500,000. He declined to say what kind of margin the company is operating on now, or what it might be when the company is profitable.

New York is Gett’s fastest growing market. It operates in the city as Juno after acquiring its competitor for $200 million a year ago, and it currently has more than 45,000 drivers on its books, or more than half of the 80,000 licensed cab drivers in the city.

Uber has been aggressive in its efforts to expand rapidly across the globe, and to position itself as a provider of all things transportation to all people — with categories like food delivery, bikes, autonomous vehicles and flying cars all in Uber’s purview. To do this, it, along with a select few others like Didi in China and Lyft in the US, have raised billions of dollars in outside funding, effectively sucking up a large part of venture money and interest in the transportation-on-demand sector.

But rather than falling by the wayside for being unable to keep up, Gett has turned the situation into an opportunity of sorts, by taking a very different, vastly more modest route. Its principle has largely been, for the last several years, to focus most of its efforts on nailing one area before considering how and where to grow.

“We are focusing just on our core service,” Waiser said. “We’d rather do that better than others than to go too broad. Uber might be able to afford [going broad], but we want everyone in New York [for example] to know that we are there if you want a much better, higher quality option.”

In the case of Gett, the company has been primarily serving consumers with rides, with a heavy emphasis on business users (other test areas of Gett’s like shuttle services and courier deliveries are still too small to note, Waiser said). The company today has 13,000 large enterprises on its books, and they account for the majority of the company’s revenues and profit. “Most of our billion dollars in revenues are coming from corporate clients,” he said.

The idea is not just to court businesses, but to provide a level of service that will help Gett grow by word of mouth, both among drivers and passengers. The company, he said, will only consider drivers with a 4.8 rating or higher in New York before a driver can work with Gett, using the driver’s previous ratings on other platforms as a marker. On the other side of the transaction, Gett’s been trying to give drivers better commissions than other services in their specific markets, and generally raising the standard for how they treat them, said Waiser. He said its growth in NYC has been done on virtually no marketing budget.

“Some people describe marketing as a tax, but once you become favourable for drivers you don’t have that tax,” he said. “Drivers choose us, without us having to spend a lot on marketing.”

Over in London, Waiser said that that the current state of competition — Uber is currently appealing a rejection by local regulators for a license to operate in the city — hasn’t had a huge impact. “We’ve seen some uplift in London because of it, but you always need to try to improve your service, rather than looking at competition. We believe Uber will stay in London, so we’re not building our business on competitors’ problems.”

Zebra Medical Vision, an Israeli medical imaging startup that uses machine and deep learning to build tools for radiologists, has raised a $30 million Series C led by health technology fund aMoon Ventures, with participation from Aurum, Johnson & Johnson Innovation—JJDC Inc. (the conglomerate’s venture capital arm), Intermountain Health and artificial intelligence experts Fei-Fei Li and Richard Socher. Existing investors Khosla Ventures, Nvidia, Marc Benioff, OurCrowd and Dolby Ventures also returned for the round.

Zebra also announced its Textray research today, which it claims is the “most comprehensive AI research conducted on chest X-rays to date.” Textray is being used to develop a new product that has already been trained using almost two million images to identify 40 clinical findings. Scheduled to launch next year, it will help automate the analysis of chest X-rays for radiologists.

Founder and CEO Elad Benjamin, who launched Zebra with Eyal Toledano and Eyal Gura in 2014, told TechCrunch in an email that its Series C capital will be spent on new hires to speed up the development of its analytics engine and its go-to-market strategy.

Chest X-rays are one of the most common radiographs ordered, but also among the most difficult for radiologists to interpret. There are several groups of researchers focused on using artificial intelligence algorithms to make the process more accurate and efficient, including teams from Google, Stanford and the Dubai Health Authority.

Benjamin said Zebra, which has now raised a total of $50 million, differentiates its approach by looking at its algorithms from a “holistic product perspective. Developing an algorithm is just one piece,” he added. “Integrating it into the workflow, adapting it to the needs of multiple countries and healthcare environments, supporting and updating it, and regulating it properly globally takes a tremendous focus – and that’s what delivers value, over and above an algorithm.”

He added that “it will take a few more years” before AI becomes mainstream in medical imaging and diagnosis, but he believes that it will eventually be a critical component of radiology. Zebra Medical Vision already markets a bundle of algorithms for lung, breast, liver, cardiovascular and bone disease diagnoses called AI1 that costs hospitals $1 per scan.

In a press statement, aMoon managing partner Dr. Yair Schindel said “We are excited to partner with the Zebra team, which is harnessing the power of data and machine learning to provide physicians and healthcare systems with tools to dramatically increase capacity, while improving patient care. This investment aligns with our vision of backing scalable and sustainable innovations that will have a valuable impact on fundamental facets of global healthcare.”

Zebra Medical Vision, an Israeli medical imaging startup that uses machine and deep learning to build tools for radiologists, has raised a $30 million Series C led by health technology fund aMoon Ventures, with participation from Aurum, Johnson & Johnson Innovation—JJDC Inc. (the conglomerate’s venture capital arm), Intermountain Health and artificial intelligence experts Fei-Fei Li and Richard Socher. Existing investors Khosla Ventures, Nvidia, Marc Benioff, OurCrowd and Dolby Ventures also returned for the round.

Zebra also announced its Textray research today, which it claims is the “most comprehensive AI research conducted on chest X-rays to date.” Textray is being used to develop a new product that has already been trained using almost two million images to identify 40 clinical findings. Scheduled to launch next year, it will help automate the analysis of chest X-rays for radiologists.

Founder and CEO Elad Benjamin, who launched Zebra with Eyal Toledano and Eyal Gura in 2014, told TechCrunch in an email that its Series C capital will be spent on new hires to speed up the development of its analytics engine and its go-to-market strategy.

Chest X-rays are one of the most common radiographs ordered, but also among the most difficult for radiologists to interpret. There are several groups of researchers focused on using artificial intelligence algorithms to make the process more accurate and efficient, including teams from Google, Stanford and the Dubai Health Authority.

Benjamin said Zebra, which has now raised a total of $50 million, differentiates its approach by looking at its algorithms from a “holistic product perspective. Developing an algorithm is just one piece,” he added. “Integrating it into the workflow, adapting it to the needs of multiple countries and healthcare environments, supporting and updating it, and regulating it properly globally takes a tremendous focus – and that’s what delivers value, over and above an algorithm.”

He added that “it will take a few more years” before AI becomes mainstream in medical imaging and diagnosis, but he believes that it will eventually be a critical component of radiology. Zebra Medical Vision already markets a bundle of algorithms for lung, breast, liver, cardiovascular and bone disease diagnoses called AI1 that costs hospitals $1 per scan.

In a press statement, aMoon managing partner Dr. Yair Schindel said “We are excited to partner with the Zebra team, which is harnessing the power of data and machine learning to provide physicians and healthcare systems with tools to dramatically increase capacity, while improving patient care. This investment aligns with our vision of backing scalable and sustainable innovations that will have a valuable impact on fundamental facets of global healthcare.”

Restaurant food delivery startup Deliveroo is opening up Point of Sale (POS) integrations to restaurant partners, via an API and developer portal, after trialling the approach this spring and finding appetite for uptake.

The integration is intended to free up front-of-house staff from having to manually input Deliveroo food orders into the restaurant’s sales system.

A Deliveroo spokesman told us that one major chain had been having to assign a single member of staff at peak times to just “sit there and type in the orders”, thereby reducing the number of staff who could attend to customers in the restaurant at a busy time.

Order inputting also isn’t necessarily popular with staff, given there’s no direct opportunity for tips. Automating the task certainly seems a no brainer.

Deliveroo claims initial tests show the integration cuts order processing time from ~2.5 mins to ~30 sec. It also reckons POS companies and “tech savvy restaurants” can complete the integration in two weeks.

“We weren’t sure how many people were going to adopt it — now we can see it’s clearly taken off and is working supporting restaurants,” said the spokesman.

So far around 500 restaurant sites are using integration, having joined during the trial, along with 25 POS companies that are plugging into its API. And Deliveroo says it expects thousands to be doing so by the end of the year.

A “couple of hundred” restaurants are joining each month, the spokesman added.

He said Deliveroo opted to go for an API approach — and rely on restaurants to integrate its order system into their sales systems — because of how diverse the POS market is.

“At a European level it’s crazy how many different types of POS companies there are. So from our perspective, rather than trying to produce our own system which worked with every single one it was best to design a system where they were able to use an API of ours and integrate from their system,” the spokesman told us.

Deliveroo now operates in 12 markets, with eight in Europe and four further afield. And so far POS integrations have been achieved with restaurants in the UK, France, Belgium and Spain — with the UK, where it has with more than 10,000 restaurant partners, being the most “advanced” so far.

The spokesman said it is also working on rolling out integrations to all other markets “in coming months”.

“There are still some places where this isn’t going to work. And this is where firms tend to use middleware companies. And we still are working with a middleware company for places where the direct integration isn’t going to work,” he added.

“But in terms of a principal, Uber have gone down the principal of they’ve bought a middleware company and they’re going to do the middleware themselves. [But] if you can reduce your reliance on middleware you reduce the chances of errors, problems you face and an extra link in the chain that could go down and prevent orders.”

Restaurant food delivery startup Deliveroo is opening up Point of Sale (POS) integrations to restaurant partners, via an API and developer portal, after trialling the approach this spring and finding appetite for uptake.

The integration is intended to free up front-of-house staff from having to manually input Deliveroo food orders into the restaurant’s sales system.

A Deliveroo spokesman told us that one major chain had been having to assign a single member of staff at peak times to just “sit there and type in the orders”, thereby reducing the number of staff who could attend to customers in the restaurant at a busy time.

Order inputting also isn’t necessarily popular with staff, given there’s no direct opportunity for tips. Automating the task certainly seems a no brainer.

Deliveroo claims initial tests show the integration cuts order processing time from ~2.5 mins to ~30 sec. It also reckons POS companies and “tech savvy restaurants” can complete the integration in two weeks.

“We weren’t sure how many people were going to adopt it — now we can see it’s clearly taken off and is working supporting restaurants,” said the spokesman.

So far around 500 restaurant sites are using integration, having joined during the trial, along with 25 POS companies that are plugging into its API. And Deliveroo says it expects thousands to be doing so by the end of the year.

A “couple of hundred” restaurants are joining each month, the spokesman added.

He said Deliveroo opted to go for an API approach — and rely on restaurants to integrate its order system into their sales systems — because of how diverse the POS market is.

“At a European level it’s crazy how many different types of POS companies there are. So from our perspective, rather than trying to produce our own system which worked with every single one it was best to design a system where they were able to use an API of ours and integrate from their system,” the spokesman told us.

Deliveroo now operates in 12 markets, with eight in Europe and four further afield. And so far POS integrations have been achieved with restaurants in the UK, France, Belgium and Spain — with the UK, where it has with more than 10,000 restaurant partners, being the most “advanced” so far.

The spokesman said it is also working on rolling out integrations to all other markets “in coming months”.

“There are still some places where this isn’t going to work. And this is where firms tend to use middleware companies. And we still are working with a middleware company for places where the direct integration isn’t going to work,” he added.

“But in terms of a principal, Uber have gone down the principal of they’ve bought a middleware company and they’re going to do the middleware themselves. [But] if you can reduce your reliance on middleware you reduce the chances of errors, problems you face and an extra link in the chain that could go down and prevent orders.”

Laka, a London-based insurtech startup that offers what it calls “crowd insurance” to rival traditional premiums and is initially targeting high-end bicycle owners, has raised $1.5 million in seed funding. The round is led by publicly-listed Tune Protect Group, with participation from Silicon Valley’s 500 Startups — money that will be used to enter new insurance categories and for international expansion, including South East Asia.

Founded in 2017, Laka has developed what it claims is a unique insurance model that sees customers only pay for the true cost of cover. At the end of each month, the cost of any claims is split fairly between customers, with the individual’s maximum premium capped at the “market rate”. If there is no claim, the premium that month is zero. To date, the startup says it has saved customers more than 80 percent compared to market prices.

What’s interesting about this model is that it is potentially much-better aligned with customers, meaning that fewer claims mean lower costs for the entire Laka customer base. Laka itself only makes money when a claim is made — it adds 25 percent on top of each claim to cover costs and create some margin. As long as it stays on top of fraudulent claims, customers stand to benefit with a more cost-effective and fairer insurance product.

“Customers join without paying any upfront premiums. When there is a claim, we settle it with working capital we borrow from our insurance partner in exchange for a fee,” explains Laka co-founder Jens Hartwig. “At the end of the month, we total up all claims we have settled, add our fee on top, and split the bill on a pro-rata basis. Thus, we pay out first and then ask customers to pay us back the expenses incurred”.

In contrast, the more a traditional insurer pays out in claims, the less profit it makes. “It’s a great business model from the insurer’s point of view as they happily take customer’s money and maybe settle a claim down the line. In the meantime they can reinvest the available capital. This proposition is clearly not as attractive from the customer’s’ point of view,” says Hartwig.

To change this, Laka’s model moves away from “underwriting risk” to credit risk — that is, ensuring customers can pay the required, albeit capped premium when the startup does have to pay out, which Hartwig reckons is an easily manageable risk with credit cards and modern payment providers such as Stripe.

The cap — where the monthly premium has a maximum so that Laka’s customers never face bill shock — is being provided by Zurich U.K. in the form of a stop-loss agreement for which Laka pays a small fixed fee per policy, per month. Any exposure above the cap is absorbed by Zurich, acting like a reinsurer.

Hartwig says that in months with a lot of claims, this is where the stop-loss kicks in, capping each customer’s exposure at a clearly communicated level. The promise is that you will never be charged more than competitors, but — crucially — if everyone takes better care, you will pay much less.

“We effectively offer a profit share to our customers, encouraging improved behaviour as they benefit from taking better care. By changing the way we earn money in the business model, we fixed the conflict of interest between customer and insurer,” adds the Laka co-founder.

Laka, a London-based insurtech startup that offers what it calls “crowd insurance” to rival traditional premiums and is initially targeting high-end bicycle owners, has raised $1.5 million in seed funding. The round is led by publicly-listed Tune Protect Group, with participation from Silicon Valley’s 500 Startups — money that will be used to enter new insurance categories and for international expansion, including South East Asia.

Founded in 2017, Laka has developed what it claims is a unique insurance model that sees customers only pay for the true cost of cover. At the end of each month, the cost of any claims is split fairly between customers, with the individual’s maximum premium capped at the “market rate”. If there is no claim, the premium that month is zero. To date, the startup says it has saved customers more than 80 percent compared to market prices.

What’s interesting about this model is that it is potentially much-better aligned with customers, meaning that fewer claims mean lower costs for the entire Laka customer base. Laka itself only makes money when a claim is made — it adds 25 percent on top of each claim to cover costs and create some margin. As long as it stays on top of fraudulent claims, customers stand to benefit with a more cost-effective and fairer insurance product.

“Customers join without paying any upfront premiums. When there is a claim, we settle it with working capital we borrow from our insurance partner in exchange for a fee,” explains Laka co-founder Jens Hartwig. “At the end of the month, we total up all claims we have settled, add our fee on top, and split the bill on a pro-rata basis. Thus, we pay out first and then ask customers to pay us back the expenses incurred”.

In contrast, the more a traditional insurer pays out in claims, the less profit it makes. “It’s a great business model from the insurer’s point of view as they happily take customer’s money and maybe settle a claim down the line. In the meantime they can reinvest the available capital. This proposition is clearly not as attractive from the customer’s’ point of view,” says Hartwig.

To change this, Laka’s model moves away from “underwriting risk” to credit risk — that is, ensuring customers can pay the required, albeit capped premium when the startup does have to pay out, which Hartwig reckons is an easily manageable risk with credit cards and modern payment providers such as Stripe.

The cap — where the monthly premium has a maximum so that Laka’s customers never face bill shock — is being provided by Zurich U.K. in the form of a stop-loss agreement for which Laka pays a small fixed fee per policy, per month. Any exposure above the cap is absorbed by Zurich, acting like a reinsurer.

Hartwig says that in months with a lot of claims, this is where the stop-loss kicks in, capping each customer’s exposure at a clearly communicated level. The promise is that you will never be charged more than competitors, but — crucially — if everyone takes better care, you will pay much less.

“We effectively offer a profit share to our customers, encouraging improved behaviour as they benefit from taking better care. By changing the way we earn money in the business model, we fixed the conflict of interest between customer and insurer,” adds the Laka co-founder.