Sep
06
2021
--

Fractory raises $9M to rethink the manufacturing supply chain for metalworks

The manufacturing industry took a hard hit from the Covid-19 pandemic, but there are signs of how it is slowly starting to come back into shape — helped in part by new efforts to make factories more responsive to the fluctuations in demand that come with the ups and downs of grappling with the shifting economy, virus outbreaks and more. Today, a businesses that is positioning itself as part of that new guard of flexible custom manufacturing — a startup called Fractory — is announcing a Series A of $9 million (€7.7 million) that underscores the trend.

The funding is being led by OTB Ventures, a leading European investor focussed on early growth, post-product, high-tech start-ups, with existing investors Trind VenturesSuperhero CapitalUnited Angels VCStartup Wise Guys and Verve Ventures also participating.

Founded in Estonia but now based in Manchester, England — historically a strong hub for manufacturing in the country, and close to Fractory’s customers — Fractory has built a platform to make it easier for those that need to get custom metalwork to upload and order it, and for factories to pick up new customers and jobs based on those requests.

Fractory’s Series A will be used to continue expanding its technology, and to bring more partners into its ecosystem.

To date, the company has worked with more than 24,000 customers and hundreds of manufacturers and metal companies, and altogether it has helped crank out more than 2.5 million metal parts.

To be clear, Fractory isn’t a manufacturer itself, nor does it have no plans to get involved in that part of the process. Rather, it is in the business of enterprise software, with a marketplace for those who are able to carry out manufacturing jobs — currently in the area of metalwork — to engage with companies that need metal parts made for them, using intelligent tools to identify what needs to be made and connecting that potential job to the specialist manufacturers that can make it.

The challenge that Fractory is solving is not unlike that faced in a lot of industries that have variable supply and demand, a lot of fragmentation, and generally an inefficient way of sourcing work.

As Martin Vares, Fractory’s founder and MD, described it to me, companies who need metal parts made might have one factory they regularly work with. But if there are any circumstances that might mean that this factory cannot carry out a job, then the customer needs to shop around and find others to do it instead. This can be a time-consuming, and costly process.

“It’s a very fragmented market and there are so many ways to manufacture products, and the connection between those two is complicated,” he said. “In the past, if you wanted to outsource something, it would mean multiple emails to multiple places. But you can’t go to 30 different suppliers like that individually. We make it into a one-stop shop.”

On the other side, factories are always looking for better ways to fill out their roster of work so there is little downtime — factories want to avoid having people paid to work with no work coming in, or machinery that is not being used.

“The average uptime capacity is 50%,” Vares said of the metalwork plants on Fractory’s platform (and in the industry in general). “We have a lot more machines out there than are being used. We really want to solve the issue of leftover capacity and make the market function better and reduce waste. We want to make their factories more efficient and thus sustainable.”

The Fractory approach involves customers — today those customers are typically in construction, or other heavy machinery industries like ship building, aerospace and automotive — uploading CAD files specifying what they need made. These then get sent out to a network of manufacturers to bid for and take on as jobs — a little like a freelance marketplace, but for manufacturing jobs. About 30% of those jobs are then fully automated, while the other 70% might include some involvement from Fractory to help advise customers on their approach, including in the quoting of the work, manufacturing, delivery and more. The plan is to build in more technology to improve the proportion that can be automated, Vares said. That would include further investment in RPA, but also computer vision to better understand what a customer is looking to do, and how best to execute it.

Currently Fractory’s platform can help fill orders for laser cutting and metal folding services, including work like CNC machining, and it’s next looking at industrial additive 3D printing. It will also be looking at other materials like stonework and chip making.

Manufacturing is one of those industries that has in some ways been very slow to modernize, which in a way is not a huge surprise: equipment is heavy and expensive, and generally the maxim of “if it ain’t broke, don’t fix it” applies in this world. That’s why companies that are building more intelligent software to at least run that legacy equipment more efficiently are finding some footing. Xometry, a bigger company out of the U.S. that also has built a bridge between manufacturers and companies that need things custom made, went public earlier this year and now has a market cap of over $3 billion. Others in the same space include Hubs (which is now part of Protolabs) and Qimtek, among others.

One selling point that Fractory has been pushing is that it generally aims to keep manufacturing local to the customer to reduce the logistics component of the work to reduce carbon emissions, although as the company grows it will be interesting to see how and if it adheres to that commitment.

In the meantime, investors believe that Fractory’s approach and fast growth are strong signs that it’s here to stay and make an impact in the industry.

“Fractory has created an enterprise software platform like no other in the manufacturing setting. Its rapid customer adoption is clear demonstrable feedback of the value that Fractory brings to manufacturing supply chains with technology to automate and digitise an ecosystem poised for innovation,” said Marcin Hejka in a statement. “We have invested in a great product and a talented group of software engineers, committed to developing a product and continuing with their formidable track record of rapid international growth

Oct
20
2020
--

Microsoft debuts Azure Space to cater to the space industry, partners with SpaceX for Starlink data center broadband

Microsoft is taking its Azure cloud computing platform to the final frontier — space. It now has a dedicated business unit called Azure Space for that purpose, made up of industry heavyweights and engineers who are focused on space-sector services, including simulation of space missions, gathering and interpreting satellite data to provide insights and providing global satellite networking capabilities through new and expanded partnerships.

One of Microsoft’s new partners for Azure Space is SpaceX, the progenitor and major current player in the so-called “New Space” industry. SpaceX will be providing Microsoft with access to its Starlink low-latency satellite-based broadband network for Microsoft’s new Azure Modular Datacenter (MDC) — essentially an on-demand container-based data center unit that can be deployed in remote locations, either to operate on their own or boost local capabilities.

Image Credits: Microsoft

The MDC is a contained unit, and can operate off-grid using its own satellite network connectivity add-on. It’s similar in concept to the company’s work on underwater data centres, but keeping it on the ground obviously opens up more opportunities in terms of locating it where people need it, rather than having to be proximate to an ocean or sea.

The other big part of this announcement focuses on space preparedness via simulation. Microsoft revealed the Azure Orbital Emulator today, which provides in a computer emulated environment the ability to test satellite constellation operations in simulation, using both software and hardware. It’s basically aiming to provide as close to in-space conditions as are possible on the ground in order to get everything ready for coordinating large, interconnected constellations of automated satellites in low Earth orbit, an increasing need as more defense agencies and private companies pursue this approach versus the legacy method of relying on one, two or just a few large geosynchronous spacecraft.

Image Credits: Microsoft

Microsoft says the goal with the Orbital Emulator is to train AI for use on orbital spacecraft before those spacecraft are actually launched — from the early development phase, right up to working with production hardware on the ground before it takes its trip to space. That’s definitely a big potential competitive advantage, because it should help companies spot even more potential problems early on while they’re still relatively easy to fix (not the case on orbit).

This emulated environment for on-orbit mission prep is already in use by Azure Government customers, the company notes. It’s also looking for more partners across government and industry for space-related services, including communication, national security, satellite services including observation and telemetry and more.

Sep
09
2020
--

Xometry raises $75M Series E to expand custom manufacturing marketplace

When companies need to find manufacturers to build custom parts, it’s not always an easy process, especially during a pandemic. Xometry, a seven-year-old startup based in Maryland, has built an online marketplace where companies can find manufacturers across the world with excess capacity to build whatever they need. Today, the company announced a $75 million Series E investment to keep expanding the platform.

T. Rowe Price Associates led the investment, with participation from new firms Durable Capital Partners LP and ArrowMark Partners. Previous investors also joined the round, including BMW i Ventures, Greenspring Associates, Dell Technologies Capital, Robert Bosch Venture Capital, Foundry Group, Highland Capital Partners and Almaz Capital . Today’s investment brings the total raised to $193 million, according to the company.

Company CEO and co-founder Randy Altschuler says Xometry fills a need by providing a digital way of putting buyers and manufacturers together with a dash of artificial intelligence to put the right combination together. “We’ve created a marketplace using artificial intelligence to power it, and provide an e-commerce experience for buyers of custom manufacturing and for suppliers to deliver that manufacturing,” Altschuler told TechCrunch.

The kind of custom pieces that are facilitated by this platform include mechanical parts for aerospace, defense, automotive, robotics and medical devices — what Altschuler calls mission-critical parts. Being able to put companies together in this fashion is particularly useful during COVID-19 when certain regions might have been shut down.

“COVID has reinforced the need for distributed manufacturing and our platform enables that by empowering these local manufacturers, and because we’re using technology to do it, as COVID has unfolded […] and as continents have shut down, and even specific states in the United States have shut down, our platform has allowed customers to autocorrect and shift work to other locations,” he explained

What’s more, companies could take advantage of the platform to manufacture critical personal protective equipment. “One of the beauties of our platform was when COVID hit customers could come to our platform and suddenly access this tremendous amount of manufacturing capacity to produce this much-needed PPE,” he said.

Xometry makes money by facilitating the sale between the buyer and producer. They help set the price and then make money on the difference between the cost to produce and how much the buyer was willing to pay to have it done.

They have relationships with 5,000 manufacturers located throughout the world and 30,000 customers using the platform to build the parts they need. The company currently has around 350 employees, with plans to use the money to add more to keep enhancing the platform.

Altschuler says from a human perspective, he wants his company to have a diverse workforce because he never wants to see people being discriminated against for whatever reason, but he also says as a company with an international market, having a diverse workforce is also critical to his business. “The more diversity that we have within Xometry, the more we’re able to effectively market to those folks, sell to those folks and understand how they utilize technology. We’re just going to better understand our customer set as we [build a more diverse workforce],” he said.

As a Series E-stage company, Altschuler does not shy away from the IPO question. In fact, he recently brought in new CFO Jim Rallo, who has experience taking a company public. “The market that we operate in is so large, and there’s so many opportunities for us to serve both our customers and our suppliers, and we have to be great for both of them. We need capital to do that, and the public markets can be an efficient way to access that capital and to grow our business, and in the end that’s what we want to do,” he said.

Mar
03
2020
--

Honeywell says it will soon launch the world’s most powerful quantum computer

“The best-kept secret in quantum computing.” That’s what Cambridge Quantum Computing (CQC) CEO Ilyas Khan called Honeywell‘s efforts in building the world’s most powerful quantum computer. In a race where most of the major players are vying for attention, Honeywell has quietly worked on its efforts for the last few years (and under strict NDA’s, it seems). But today, the company announced a major breakthrough that it claims will allow it to launch the world’s most powerful quantum computer within the next three months.

In addition, Honeywell also today announced that it has made strategic investments in CQC and Zapata Computing, both of which focus on the software side of quantum computing. The company has also partnered with JPMorgan Chase to develop quantum algorithms using Honeywell’s quantum computer. The company also recently announced a partnership with Microsoft.

Honeywell has long built the kind of complex control systems that power many of the world’s largest industrial sites. It’s that kind of experience that has now allowed it to build an advanced ion trap that is at the core of its efforts.

This ion trap, the company claims in a paper that accompanies today’s announcement, has allowed the team to achieve decoherence times that are significantly longer than those of its competitors.

“It starts really with the heritage that Honeywell had to work from,” Tony Uttley, the president of Honeywell Quantum Solutions, told me. “And we, because of our businesses within aerospace and defense and our business in oil and gas — with solutions that have to do with the integration of complex control systems because of our chemicals and materials businesses — we had all of the underlying pieces for quantum computing, which are just fabulously different from classical computing. You need to have ultra-high vacuum system capabilities. You need to have cryogenic capabilities. You need to have precision control. You need to have lasers and photonic capabilities. You have to have magnetic and vibrational stability capabilities. And for us, we had our own foundry and so we are able to literally design our architecture from the trap up.”

The result of this is a quantum computer that promises to achieve a quantum Volume of 64. Quantum Volume (QV), it’s worth mentioning, is a metric that takes into account both the number of qubits in a system as well as decoherence times. IBM and others have championed this metric as a way to, at least for now, compare the power of various quantum computers.

So far, IBM’s own machines have achieved QV 32, which would make Honeywell’s machine significantly more powerful.

Khan, whose company provides software tools for quantum computing and was one of the first to work with Honeywell on this project, also noted that the focus on the ion trap is giving Honeywell a bit of an advantage. “I think that the choice of the ion trap approach by Honeywell is a reflection of a very deliberate focus on the quality of qubit rather than the number of qubits, which I think is fairly sophisticated,” he said. “Until recently, the headline was always growth, the number of qubits running.”

The Honeywell team noted that many of its current customers are also likely users of its quantum solutions. These customers, after all, are working on exactly the kind of problems in chemistry or material science that quantum computing, at least in its earliest forms, is uniquely suited for.

Currently, Honeywell has about 100 scientists, engineers and developers dedicated to its quantum project.

Feb
05
2020
--

Calling all cosmic startups — pitch at TechCrunch’s space event in LA

Founders — it’s time to shoot for the stars. For the first time ever, TechCrunch is hosting TC Sessions: Space 2020 on June 25th in Los Angeles. But that’s not all, because on June 24th, TechCrunch will host a Pitch Night exclusively for early-stage space startups.

Yep, that’s right. On top of a packed programming day with fireside chats, breakout sessions and Q&As featuring the top experts and game changers in space, TechCrunch will select 10 startups focused on any aspect of space — whether you’re launching rockets, building the next big satellite constellation, translating space-based data into usable insights or even building a colony on the Moon. If your company is all about the new space startup race, and you are early stage, please apply. 

Step 1: Apply to pitch by May 15th. TechCrunch’s editorial team will review all applications and select 10 companies. Founders will be notified by June 7th.  

You’ll pitch your startup at a private event in front of TechCrunch editors, main-stage speakers and industry experts. Our panel of judges will select five finalists to pitch onstage at TC Sessions: Space. 

You will be pitching your startup to the most prestigious, influential and expert industry leaders, and you’ll get video coverage on TechCrunch, too! And the final perk? Each of the 10 startup teams selected for the Pitch Night will be given two free tickets to attend TC Sessions: Space 2020. Shoot your shot — apply here.

Even if you’re not necessarily interested in pitching, grab your ticket for a front-row seat to this event for the early-bird price of $349. If you are interested in bringing a group of five or more from your company, you’ll get an automatic 20% discount. We even have discounts for the government/military, nonprofit/NGOs and students currently attending university. Grab your tickets at these reduced rates before prices increase.

Is your company interested in sponsoring or exhibiting at TC Sessions: Space 2020? Contact our sponsorship sales team by filling out this form.

Dec
03
2019
--

OrbitsEdge partners with HPE on orbital data center computing and analytics

What kinds of businesses might be able to operate in space? Well, data centers are one potential target you might not have thought of. Space provides an interesting environment for data center operations, including advanced analytics operations and even artificial intelligence, due in part to the excellent cooling conditions and reasonable access to renewable power supply (solar). But there are challenges, which is why a new partnership between Florida-based space startup OrbitsEdge and Hewlett Packard Enterprises (HPE) makes a lot of sense.

The partnership will make OrbitsEdge a hardware supplier for HPE’s Edgeline Converged Edge Systems, and basically it means that the space startup will be handling everything required to “harden” the standard HPE micro-data center equipment for use in outer space. Hardening is a standard process for getting stuff ready to use in space, and essentially prepares equipment to withstand the increased radiation, extreme temperatures and other stressors that space adds to the mix.

OrbitsEdge, founded earlier this year, has developed a proprietary piece of hardware called the “SatFrame” which is designed to counter the stress of a space-based operating environment, making it relatively easy to take off-the-shelf Earth equipment like the HPE Edgeline system and get it working in space without requiring a huge amount of additional, custom work.

In terms of what this will potentially provide, the partnership will mean it’s more feasible than ever to set up a small-scale data center in orbit to handle at least some of the processing of space-based data right near where it’s collected, rather than having to shuttle it back down to Earth. That process can be expensive, and difficult to source in terms of even finding companies and infrastructure to use. As with in-space manufacturing, doing things locally could save a lot of overhead and unlock tons of potential down the line.

Oct
11
2019
--

Descartes Labs snaps up $20M more for its AI-based geospatial imagery analytics platform

Satellite imagery holds a wealth of information that could be useful for industries, science and humanitarian causes, but one big and persistent challenge with it has been a lack of effective ways to tap that disparate data for specific ends.

That’s created a demand for better analytics, and now, one of the startups that has been building solutions to do just that is announcing a round of funding as it gears up for expansion. Descartes Labs, a geospatial imagery analytics startup out of Santa Fe, New Mexico, is today announcing that it has closed a $20 million round of funding, money that CEO and founder Mark Johnson described to me as a bridge round ahead of the startup closing and announcing a larger growth round.

The funding is being led by Union Grove Venture Partners, with Ajax Strategies, Crosslink Capital, and March Capital Partners (which led its previous round) also participating. It brings the total raised by Descartes Labs to $60 million, and while Johnson said the startup would not be disclosing its valuation, PitchBook notes that it is $220 million ($200 million pre-money in this round).

As a point of comparison, another startup in the area of geospatial analytics, Orbital Insight, is reportedly now raising money at a $430 million valuation (that data is from January of this year, and we’ve contacted the company to see if it ever closed).

Santa Fe — a city popular with retirees that counts tourism as its biggest industry — is an unlikely place to find a tech startup. Descartes Labs’ presence there is a result of that fact that it is a spinoff from the Los Alamos National Laboratory near the city.

Johnson — who had lived in San Francisco before coming to Santa Fe to help create Descartes Labs (his previous experience building Zite for media, he said, led the Los Alamos scientists to first conceive of the Descartes Labs IP as the basis of a kind of search engine) — admitted that he never thought the company would stay headquartered there beyond a short initial phase of growth of six months.

However, it turned out that the trends around more distributed workforces (and cloud computing to enable that), engineers looking for employment alternatives to living in pricey San Francisco, plus the heated competition for talent you get in the Valley all came together in a perfect storm that helped Descartes Labs establish and thrive on its home turf.

Descartes Labs — named after the seminal philosopher/mathematician Rene Descartes — describes itself as a “data refinery”. By this, it means it injests a lot of imagery and unstructured data related to the earth that is picked up primarily by satellites but also other sensors (Johnson notes that its sources include data from publicly available satellites; data from NASA and the European space agency, and data from the companies themselves); applies AI-based techniques including computer vision analysis and machine learning to make sense of the sometimes-grainy imagery; and distills and orders it to create insights into what is going on down below, and how that is likely to evolve.

Screenshot 2019 10 11 at 13.26.33

This includes not just what is happening on the surface of the earth, but also in the air above it: Descartes Labs has worked on projects to detect levels of methane gas in oil fields, the spread of wildfires, and how crops might grow in a particular area, and the impact of weather patterns on it all.

It has produced work for a range of clients that have included governments (the methane detection, pictured above, was commissioned as part of New Mexico’s effort to reduce greenhouse gas emissions), energy giants and industrial agribusiness, and traders.

“The idea is to help them take advantage of all the new data going online,” Johnson said, noting that this can help, for example, bankers forecast how much a commodity will trade for, or the effect of a change in soil composition on a crop.

The fact that Descartes Labs’ work has connected it with the energy industry gives an interesting twist to the use of the phrase “data refinery”. But in case you were wondering, Johnson said that the company goes through a process of vetting potential customers to determine if the data Descartes Labs provides to them is for a positive end, or not.

“We have a deep belief that we can help them become more efficient,” he said. “Those looking at earth data are doing so because they care about the planet and are working to try to become more sustainable.”

Johnson also said (in answer to my question about it) that so far, there haven’t been any instances where the startup has been prohibited to work with any customers or countries, but you could imagine how — in this day of data being ‘the new oil’ and the fulcrum of power — that could potentially be an issue. (Related to this: Orbital Insight counts In-Q-Tel, the CIA’s venture arm, as one of its backers.)

Looking ahead, the company is building what it describes as a “digital twin” of the earth, the idea being that in doing so it can better model the imagery that it injests and link up data from different regions more seamlessly (since, after all, a climatic event in one part of the world inevitably impacts another). Notably, “digital twinning” is a common concept that we see applied in other AI-based enterprises to better predict activity: this is the approach that, for example, Forward Networks takes when building models of an enterprise’s network to determine how apps will behave and identify the reasons behind an outage.

In addition to the funding round, Descartes Labs named Phil Fraher its new CFO, and is announcing Veery Maxwell, Director for Energy Innovation and Patrick Cairns, who co-founded UGVP, as new board observers.

Powered by WordPress | Theme: Aeros 2.0 by TheBuckmaker.com