Virtual health and wellness platforms have grown increasingly popular throughout the pandemic, but a new startup wants to focus that effort exclusively on senior citizens. Bold, a digital health and wellness service, plans to prevent chronic health problems in older adults through free and personalized exercise programs. Co-founded by Amanda Rees and her partner Hari Arul, Bold picked up $7 million this week in seed funding led by Julie Yoo of Silicon Valley-based Andreessen Horowitz.

Rees said in an interview that the idea for Bold came from time she spent caring for her grandmother, helping her through health challenges like falls. “I kept thinking about solutions we could build to keep someone healthier longer, rather than waiting for until they have a fall or something else goes off the rails to intervene,” she said. Rees started Bold to use what she’d learned from her own experience in dance and yoga to help her grandmother practice maintaining balance to prevent future falls. “My passion really was around ways to sort of widen the aperture and make these solutions more accessible and built for older people.”

The member experience is pretty straightforward. Users fill out some brief fitness information on the web-based platform, outlining their goals and current baseline. From that information, Bold creates a personalized program that ranges from a short, seated Tai Chi class once a week, to cardio and strength classes meeting multiple times each week. “The idea is to really meet a member where they are, and then through our programming, help them along their journey of doing the types of exercises that are going to have the most immediate benefit for them,” said Rees.

Bold’s funding round comes at a time of concern around ballooning healthcare expenses for older populations, and a focus on how to reduce these costs for both current and future generations. While falls alone aren’t necessarily complex medical incidents, they have the potential to lead to fractures and other serious injuries. Bold’s preventative approach to falls is a more active solution than necklace or bracelet monitors that send a signal to emergency services when they detect a fall. And by offering virtual programs, they can help at-risk older populations engage in exercise while avoiding potential COVID-19 exposure at gyms.

Research shows that this works. Even simple, low-intensity exercise can improve balance and strength enough to reduce the incidence of falls, which is currently the leading cause of injury and injury death among older adults.

Fewer injuries would mean less need for medical care, which would lead to money saved for hospitals and health insurers alike. That’s why in addition to their seed funding, Bold has plans to start rolling out partnerships with Medicare Advantage organizations and risk-bearing providers, which will help make their exercise programs available to users for free.

The recent Databricks funding round, a $1 billion investment at a $28 billion valuation, was one of the year’s most notable private investments so far.

For Databricks signaled its IPO readiness by disclosing to TechCrunch last year that it had scaled its revenue run rate from $200 million to $350 million in a year, so the new capital looked like the capstone on its private fundraising before an eventual public debut.

The Exchange explores startups, markets and money. Read it every morning on Extra Crunch, or get The Exchange newsletter every Saturday.

But I did have a few questions, starting with the price of the round.

At a $28 billion valuation and ARR of $425 million, Databricks is valued at around 66x top line. That’s steep, if not the highest number we can dredge up on the public markets. Of course, for Databricks shareholders, seeing the value of their stock rise so quickly is hardly a bad thing. They are hardly going to complain about having more paper wealth.

But what about the investor perspective? Does the price really make sense? The Exchange caught up with Battery Ventures’ Dharmesh Thakker earlier this week to discuss a number of things, one of which was Databricks’ round and pricing. Thakker is named in the Databricks Series D funding announcement, which brought Battery into the company.

What was surprising about our conversation was not that Thakker was bullish on Databricks — a company that he and his firm have backed since its $140 million, 2017 round when the company was worth just under $1 billion. What surprised me was that he thinks its new $28 billion valuation might be a little low.

Intriguing, yeah? So this morning for both of us, I’ve pulled out quotes from our chat to help explain how Thakker views the market for Databricks, unicorns at scale more broadly through the lens of risk-adjusted investing, and the scale of the market some unicorns are playing in.

At the close, we’ll remind ourselves what Databricks CEO Ali Ghodsi told TechCrunch when we asked him the same question. Let’s go!

Databricks at $28 billion

Here’s how the valuation part of my chat with the Battery Ventures’ investor went down:

The Exchange: I want to talk about Databricks, because I spoke to [CEO] Ali [Ghodsi] yesterday about this round, and hot damn, it’s a lot of money at a valuation that is roughly 64x ARR, give or take. I don’t understand the price, and I know it’s a boring thing to talk about. [It’s a] great company, I get their market, I’ve talked to them a bunch, I know their revenue numbers. [But] I don’t understand the price, and I was hoping you could tell me why I’m being too conservative.

Dharmesh Thakker:  I, for what it’s worth, think [the price] fair. If anything, I think it is on the lower end — he could have done better, frankly. But I think it comes down to three major things, right?

One is the addressable market. Just think about the addressable market of data. If there’s a trillion dollars spent in software or technology, I think you and I would be both hard pressed to say, almost all of that [isn’t] influenced by some data-oriented decisioning. Whether it’s digital transformation, whether it’s analytics, data is everywhere. So the TAM is massive … I think you and I both agree on that, whether it is $20 billion or $80 billion — it’s massive.

Fundraising is challenging, especially for deep tech founders who need to get investors excited about a complex technology, a complex sales cycle and a complex risk profile.

As a former investor and current angel investor, I have met thousands of founders, many in the deep tech space.

Based on my experience, here’s how to avoid making the most common mistakes deep tech founders make when pitching investors:

Work on your storytelling

Highlight your big vision

Early-stage investors are in the business of funding dreams. They chose to be early-stage investors because they love hearing about new ideas and enthralling futures. They deliberately are not investment bankers or accountants because they do not want to constantly pour over endless spreadsheets or dive deep into financial models. Similarly, they are not operators because they do not want to spend time figuring out the intricacies of a supply chain or a marketing campaign or the configuration of a product component.

So make your pitch tailored to what excites venture capital investors and avoid what does not. Keep the financial model details and the warehouse system logistics information to your Appendix. You have it in case anyone wants to dive in deeper, but your core presentation should be focused on your biggest, most bullish hopes for the company seven to 10 years from now. Dedicate multiple slides to painting the picture of what society would look like should you meet all your intended milestones as a company.

Underscore the impact

As a deep tech company, your differentiation is in your intellectual property. However, investors care less about the “what” and much more about the “so what.” Investors are less interested in the intricacies of your technology and more interested in what impact it can create.

Formulate your slides to focus on answering questions like, “What can people or companies do as a result of your technology?” and “How will people save time, money and lives with your product?”

Put your presentation to the “grandma” test. Would your grandmother be able to understand and be excited about everything you share? Investor pitch meetings are not dissertation defenses. You are being evaluated on your potential for impact rather than the intricate details of your research. The best way to succeed in this evaluation framework is to ensure that everything you share is relevant and exciting to a diverse audience of even nontechnical folks.

Try to reach hearts and minds

Five million people are a statistic, but one person is a story. When people read data on massive populations of people, they conceptually understand the implications but only on a logical level, not an emotional one. When pitching, you want to reach the hearts of investors.

Meet moka.care, a French startup that has built several services that should help you improve your psychological well-being. The company sells its solution to employers directly. They can then offer access to moka.care to their employees.

The startup raised a $3 million (€2.5 million) funding round from Singular, the VC firm founded by former Alven partners Jeremy Uzan and Raffi Kamber. A long list of business angels are also participating in today’s round, such as Nicolas Dessaigne (Algolia), Ning Li (Made.com, Typology), Florian Douetteau (Dataiku), Céline Lazorthes (Leetchi, MangoPay), Pierre Dubuc (OpenClassrooms), Marc-Antoine de Longevialle (LeCab), Adrien Ledoux (JobTeaser), Roxanne Varza (Station F), Thibault Lamarque (CASTALIE) and Côme Fouques (Indy).

Moka.care believes that companies aren’t doing enough when it comes to mental health. Many companies give you a phone number and tell you that you can call that number to get mental support. But few employees actually call those helplines.

That’s why the startup is taking a completely different approach. The most important principle is that people are looking for different things. And you don’t necessarily know what you’re looking for when you’re feeling down. When you first contact moka.care, the company spends roughly half an hour talking with you to understand what you’re looking for.

There are three main options after that. Moka.care could send you some recommendations for a practitioner — it can be a psychologist, a certified coach or a licensed therapist. Moka.care also organizes group sessions around a specific topic. It could be focused on remote work, work-life balance, self-confidence, etc. Finally, moka.care also provides content on some of those topics. You can access that content and learn more about yourself.

With this granular approach, the company hopes it can tackle mental health conditions before it’s too late — you don’t want to recommend a therapist when an employee is already suffering from excessive stress, fatigue or burnout.

Employees don’t pay for the first sessions as it’s part of moka.care’s plans. This way, the barrier to entry should be much lower for employees. Of course, if you want to book further appointments, you’ll have to pay at some point.

For employers, moka.care tries to lower the barrier to entry as well. Clients agree on a per-employee-per-month subscription plan based on some usage rate. If your employees end up using moka.care more than that, you don’t pay more. If your employees don’t use the service at all and you’re overpaying, the startup pays you back.

There are 30 companies currently using moka.care — it represents thousands of employees that could potentially create an account and access the service. The startup currently works with around 50 practitioners.

Flux, the London fintech that has built a technology platform for banks and merchants to power itemised digital receipts and more, has seen its lengthy pilot with Barclays bear fruit.

Announced formally today — but actually quietly rolled out a few months ago — Flux-powered digital receipts are now available as an opt-in for all U.K. Barclays debit card holders within the bank’s main mobile banking app. Previously, the functionality was only available within the Barclays Launchpad app, which is available for customers that want to try out experimental or upcoming features.

Early last year, Barclays announced that it has invested in Flux, taking a minority stake, so the strengthening of its partnership isn’t too much of a surprise. Flux also went through the Techstars-powered Barclays accelerator in its very early days. However, not all corporate accelerators lead to great outcomes as corporates are notoriously risk-adverse. This one certainly wasn’t rushed but it’s meaningful regardless, giving Flux a major shot in the arm in reaching mainstream banking customers beyond the existing challenger bank partnerships it has forged.

“Customers who pay using their Barclays debit card for future in store purchases at H&M, shoe retailer schuh and food outlets, which include Just Eat and Papa Johns, will see their receipts sent automatically to their app after making a purchase. They can then easily and securely view their receipts whenever they need by tapping on the transaction,” says Barclays. Crucially, although opt-in, Barclays customers will receive a prompt to set up digital receipts when they purchase items from retailers currently on-boarded to Flux.

Founded in 2016 by former early employees at Revolut, Flux bridges the gap between the itemised receipt data captured by a merchant’s point-of-sale (POS) system and what little information typically shows up on your bank statement or mobile banking app. Off the back of this, it can also power loyalty schemes and card-linked offers, as well as give merchants much deeper POS analytics via aggregated and anonymised data on consumer behaviour, such as which products are selling best in unique baskets.

On the banking side, along with Barclays, Flux has partnered with challenger banks Starling and Monzo. Once banking customers link their account to the service, Flux delivers digital receipts (and where available rewards and loyalty) for transactions at Flux retailer partners.

Longer term, Flux wants to become a standard for the interchange of item level digital receipt data — and the proprietary platform that powers that standard — but has always faced a chicken-and-egg problem: It needs bank integrations to sign up merchants and it needs merchant integrations to sign up banks. Barclays going live properly is another significant turn in the upstart’s flywheel.

Latitude, a startup building games with “infinite storylines” generated by artificial intelligence, is announcing that it has raised $3.3 million in seed funding.

The idea of an AI-generated story might make you think of hilariously nonsensical experiments like “Sunspring,” but Latitude’s first title, AI Dungeon, is an impressively open-ended (and coherent) text adventure game where you can choose from a wide variety of genres and characters.

Unlike a classic text adventure like Zork — where players quickly become familiar with “you can’t do that”-style messages when they type something the designers hadn’t planned for — AI Dungeon can respond to any command. For example, when my brave knight was charging into battle, I typed “get depressed” and he quickly sat on a rock with his head between his hands.

“How does the AI know what’s a good story?” said co-founder and CEO Nick Walton. “Because it’s read a lot of good stories and knows the patterns involved in that.”

AI Dungeon actually started out as one of Walton’s hackathon projects. While the initial version didn’t win any prizes, he kept at it, assisted by improvements in OpenAI’s language generator, of which the most recent version is GPT-3.

“The very first version of AI Dungeon I built was coherent on a sentence level, but on a paragraph level it made no sense,” Walton said. “Once you get to GPT-2, it makes a lot more sense. Once you get to GPT-3, it’s a lot more coherent on a story level. And so I think to a degree, these issues with coherency, the story not making sense, get solved as the AI gets better.”

Latitude says AI Dungeon is attracting 1.5 million monthly active users. The startup plans to create more AI-powered games, and eventually to release a platform allowing other game designers to do the same.

Walton noted that without AI, video games are always constrained by the imagination of its creators. Even when you get to games like The Elder Scrolls II: Daggerfall or No Man’s Sky, with randomly generated towns or planets, he argued that they’re really offering “the same spin on a similar concept.”

For example, he said that in Daggerfall, “When you go to all these towns, they’re all basically the same. That’s the problem with procedural generation: You’re not coming up with unique things.” AI, on the other hand, can come up with “something completely unique that’s so, so different every time.”

From a business perspective, he said that this could lower the cost of developing AAA games from more than $100 million to less than $100,000 — though Latitude has a ways to go before it reaches that level, since it hasn’t even released a game with graphics yet. Walton also said this could lead to new levels of immersion and interactivity.

“With this technology, you could have a world with tens of thousands of characters with their own hopes and wants and dreams,” he said. “You can have worlds that are dynamic, that are alive, rather than something like World of Warcraft, where you’ve got 10 million people who are doing the same quest.”

The startup’s funding was led by NFX, with participation from Album VC and Griffin Gaming Partners.

“Latitude is revolutionizing how games are made, creating a whole new genre of entertainment gaming fueled by AI,” said James Currier of NFX in a statement. “The best AI minds and engineers are gathering there to produce games that the world has never seen before. Latitude is already by far the leading AI games company.“

As impressive as the cameras in our smartphones are, they’re fundamentally limited by the physical necessities of lenses and sensors. Metalenz skips over that part with a camera made of a single “metasurface” that could save precious space and battery life in phones and other devices… and they’re about to ship it.

The concept is similar to, but not descended from, the “metamaterials” that gave rise to flat beam-forming radar and lidar of Lumotive and Echodyne. The idea is to take a complex 3D structure and accomplish what it does using a precisely engineered “2D” surface — not actually two-dimensional, of course, but usually a plane with features measured in microns.

In the case of a camera, the main components are of course a lens (these days it’s usually several stacked), which corrals the light, and an image sensor, which senses and measures that light. The problem faced by cameras now, particularly in smartphones, is that the lenses can’t be made much smaller without seriously affecting the clarity of the image. Likewise sensors are nearly at the limit of how much light they can work with. Consequently, most of the photography advancements of the last few years have been done on the computational side.

Using an engineered surface that does away with the need for complex optics and other camera systems has been a goal for years. Back in 2016 I wrote about a NASA project that took inspiration from moth eyes to create a 2D camera of sorts. It’s harder than it sounds, though — usable imagery has been generated in labs, but it’s not the kind of thing that you take to Apple or Samsung.

Metalenz aims to change that. The company’s tech is built on the work of Harvard’s Federico Capasso, who has been publishing on the science behind metasurfaces for years. He and Rob Devlin, who did his doctorate work in Capasso’s lab, co-founded the company to commercialize their efforts.

“Early demos were extremely inefficient,” said Devlin of the field’s first entrants. “You had light scattering all over the place, the materials and processes were non-standard, the designs weren’t able to handle the demands that a real world throws at you. Making one that works and publishing a paper on it is one thing, making 10 million and making sure they all do the same thing is another.”

Their breakthrough — if years of hard work and research can be called that — is the ability not just to make a metasurface camera that produces decent images, but to do it without exotic components or manufacturing processes.

“We’re really using all standard semiconductor processes and materials here, the exact same equipment — but with lenses instead of electronics,” said Devlin. “We can already make a million lenses a day with our foundry partners.”

The first challenge is more or less contained in the fact that incoming light, without lenses to bend and direct it, hits the metasurface in a much more chaotic way. Devlin’s own PhD work was concerned with taming this chaos.

“Light on a macro [i.e. conventional scale, not close-focusing] lens is controlled on the macro scale, you’re relying on the curvature to bend the light. There’s only so much you can do with it,” he explained. “But here you have features a thousand times smaller than a human hair, which gives us very fine control over the light that hits the lens.”

Those features, as you can see in this extreme close-up of the metasurface, are precisely tuned cylinders, “almost like little nano-scale Coke cans,” Devlin suggested. Like other metamaterials, these structures, far smaller than a visible or near-infrared light ray’s wavelength, manipulate the radiation by means that take a few years of study to understand.

The result is a camera with extremely small proportions and vastly less complexity than the compact camera stacks found in consumer and industrial devices. To be clear, Metalenz isn’t looking to replace the main camera on your iPhone — for conventional photography purposes the conventional lens and sensor are still the way to go. But there are other applications that play to the chip-style lens’s strengths.

Something like the FaceID assembly, for instance, presents an opportunity. “That module is a very complex one for the cell phone world — it’s almost like a Rube Goldberg machine,” said Devlin. Likewise the miniature lidar sensor.

At this scale, the priorities are different, and by subtracting the lens from the equation the amount of light that reaches the sensor is significantly increased. That means it can potentially be smaller in every dimension while performing better and drawing less power.

Lest you think this is still a lab-bound “wouldn’t it be nice if” type device, Metalenz is well on its way to commercial availability. The $10 million Series A they just raised was led by 3M Ventures, Applied Ventures LLC, Intel Capital, M Ventures and TDK Ventures, along with Tsingyuan Ventures and Braemar Energy Ventures — a lot of suppliers in there.

Unlike many other hardware startups, Metalenz isn’t starting with a short run of boutique demo devices but going big out of the gate.

“Because we’re using traditional fabrication techniques, it allows us to scale really quickly. We’re not building factories or foundries, we don’t have to raise hundreds of mils; we can use what’s already there,” said Devlin. “But it means we have to look at applications that are high volume. We need the units to be in that tens of millions range for our foundry partners to see it making sense.”

Although Devlin declined to get specific, he did say that their first partner is “active in 3D sensing” and that a consumer device, though not a phone, would be shipping with Metalenz cameras in early 2022 — and later in 2022 will see a phone-based solution shipping as well.

In other words, while Metalenz is indeed a startup just coming out of stealth and raising its A round… it already has shipments planned on the order of tens of millions. The $10 million isn’t a bridge to commercial viability but short-term cash to hire and cover upfront costs associated with such a serious endeavor. It’s doubtful anyone on that list of investors harbors any serious doubts on ROI.

The 3D sensing thing is Metalenz’s first major application, but the company is already working on others. The potential to reduce complex lab equipment to handheld electronics that can be fielded easily is one, and improving the benchtop versions of tools with more light-gathering ability or quicker operation is another.

Though a device you use may in a few years have a Metalenz component in it, it’s likely you won’t know — the phone manufacturer will probably take all the credit for the improved performance or slimmer form factor. Nevertheless, it may show up in teardowns and bills of material, at which point you’ll know this particular university spin-out has made it to the big leagues.

Werewolf: The Apocalypse -- Earthblood does a good job of making your feel like a powerful werewolf, even if the stealth isn't that good.Read More

IBM predicts commercial applications for quantum computing within the next 5 years as hardware continues to advance.Read More

Sony's expecting its most profitable year ever for PlayStation, and that is despite traditionally losing money during console launches.Read More

Kite, the ML-based code completion assistant, launched an enterprise-grade Team Server today to accelerate developer teams' workflows.Read More

Gearbox Entertainment CEO Randy Pitchford said that selling his company will be good for both gamers and employees.Read More

Hitman 3's Dartmoor stage is filled with mystery, but while you might have solved it one way, multiple other possibilities exist as well.Read More

Researchers at Tilburg University and the University of Maryland claim that AI-translated text is less 'rich' than human translations.Read More

Linux Foundation takes control of open source Magma project from Facebook to advance adoption of 5G wireless networks.Read More

Microsoft today launched Custom Neural Voice in limited preview, a service that allows customers to create custom voices using AI.Read More

Sony is selling PS5 hardware at a loss and component costs could rise as memory and processor demand remains high.Read More

Embedded finance — the idea of offering financial products where customers are already congregating via white label solutions and APIs – isn’t an entirely new concept. In fact, in one form or another, such as point of sale credit, the concept has existed for years and long before Silicon Valley venture capital firm and media company (ha!) Andreessen Horowitz made it a thing. However, fuelled by cloud technology and a plethora of new fintech and Banking-as-a-Service startups, there is no doubt the embedded finance trend is accelerating.

The latest company to declare its hand is Berlin-based Banxware, which offers embedded finance in the form of loans for SMEs, in partnership with marketplaces, payments providers, and others. It launched in December and today is disclosing that it has raised €4 million in seed funding.

Leading the round is Force over Mass, and VR Ventures. They are joined by HTGF, and private investors in banking, payment and e-commerce.

Banxware says it will use the investment to develop and grow its embedded white label financial services offering, and expand its team. In addition to lending, the startup will also soon offer card-based products and other financial services.

Banxware’s tech and infrastructure enables any company to offer loans and other banking services to SME customers. The idea is to act as the link between banks (lenders), digital platforms, and merchants. Banks get access to hard to reach SME customers. Platforms, such as online marketplaces, can up-sell financial products beyond their core offering. And merchants benefit from speedy access to working capital.

“SMEs have a hard time to access capital when needed, especially when they are less than three years old or do not have the most pristine credit history,” explains co-founder and CEO Jens Röhrborn. “On top of this, loan applications, i.e. loan decisions and loan payout, still take several weeks in most cases.

“More and more sellers and merchants are using digital platforms through which they sell their products or process their digital payments. By using the recent historic data on these merchants provided by the platforms, we can lend against their future revenues”.

This has seen Banxware build an instant lending tool that includes AML and KYC compliance, and a scoring engine that analyzes historic platform data and data from third party providers, such as account information providers and external scoring services. The promise is an instant loan decision and loan payout, “all in less than 15 minutes”.

“On the lending side, we work with both balance sheet lenders and lending vehicles with whom we pre-agree on lending terms and loan decision criteria and on whose behalf we execute the loan decision,” says Röhrborn. “Merchants repay their loan in such a way that platforms subtract a certain percentage of the future merchant payouts”.

Röhrborn says the company’s instant lending tool is “only the beginning” and that Banxware will develop additional embedded financial services and expand internationally.

Meanwhile, the German fintech currently generates revenue by charging a one time fee for each loan that is processed through its platform and via a one off customization fee.

Time is critical for healthcare providers, especially in the middle of the pandemic. Singapore-based Bot MD helps save time with an AI-based chatbot that lets doctors look up important information from their smartphones, instead of needing to call a hospital operator or access its intranet. The startup announced today it has raised a $5 million Series A led by Monk’s Hill Ventures.

Other backers include SeaX, XA Network and SG Innovate, and angel investors Yoh-Chie Lu, Jean-Luc Butel and Steve Blank. Bot MD was also part of Y Combinator’s summer 2018 batch.

The funding will be used to expand in the Asia-Pacific region, including Indonesia, the Philippines, Malaysia and Indonesia, and to add new features in response to demand from hospitals and healthcare organizations during COVID-19. Bot MD’s AI assistant currently supports English, with plans to release Bahasa Indonesian and Spanish later this year. It is currently used by about 13,000 doctors at organizations including Changi General Hospital, National University Health System, National University Cancer Institute of Singapore, Tan Tock Seng Hospital, Singapore General Hospital, Parkway Radiology and the National Kidney Transplant Institute.

Co-founder and chief executive officer Dorothea Koh told TechCrunch that Bot MD integrates hospital information usually stored in multiple systems and makes it easier to access.

Image Credits: Bot MDWithout Bot MD, doctors may need to dial a hospital operator to find which staffers are on call and get their contact information. If they want drug information, that means another call to the pharmacy. If they need to see updated guidelines and clinical protocols, that often entails finding a computer that is connected to the hospital’s intranet.

“A lot of what Bot MD does is to integrate the content that they need into a single interface that is searchable 24/7,” said Koh.

For example, during COVID-19, Bot MD introduced a new feature that takes healthcare providers to a form pre-filled with their information when they type “record temperature” into the chatbot. Many were accessing their organization’s intranet twice a day to log their temperature and Koh said being able to use the form through Bot MD has significantly improved compliance.

The time it takes to onboard Bot MD varies depending on the information systems and amount of content it needs to integrate, but Koh said its proprietary natural language processing chat engine makes training its AI relatively quick. For example, Changi General Hospital, a recent client, was onboarded in less than 10 days.

Bot MD plans to add new clinical apps to its platform, including ones for electronic medical records (EMR), billing and scheduling integrations, clinical alerts and chronic disease monitoring.

When you want to buy a refrigerator or a television, you can walk to the nearby electronics store or visit an e-commerce website like Amazon. But where do you go when you’re looking for parts of a crane, a door or chassis of different machines?

For several businesses globally, the answer to that question is increasingly Zetwerk, a Bangalore-based startup.

The three-year-old startup runs a business-to-business marketplace for manufacturing items that connects OEMs (original equipment manufacturers) and EPC (engineering procurement construction) customers with manufacturing small-businesses and enterprises.

All the products it sells today are custom-made. “Nobody has a stock of such inventories. You get the order, you find manufacturers and workshops that make them,” explained Amrit Acharya, co-founder and chief executive of Zetwerk, in an interview with TechCrunch.

Its customers — there are over 250 of them, up from 100 a year ago — operate across two-dozen industries (including process plants, oil & gas, steel, aerospace, medical devices, apparel and luxury goods) in the infrastructure space, and approach Zetwerk with digital designs they wish to be translated into physical products.

Customers aren’t alone in seeing value in Zetwerk. On Wednesday, the Indian startup said it has raised $120 million in a Series D financing round led by existing investors Greenoaks Capital and Lightspeed Venture Partners. Existing investors Sequoia Capital and Kae Capital also participated in the Series D round.

The new round, which brings Zetwerk’s to-date raise to $193 million, gives the firm a post-money valuation of somewhere between $600 million to $700 million, a person familiar with the matter told TechCrunch. (A quick side note: Zetwerk announced a $21 million Series C round last year, but ended up raising $31 million in that round.)

Zetwerk was co-founded by Acharya, Srinath Ramakkrushnan, Rahul Sharma and Vishal Chaudhary. Long before Acharya and Ramakkrushnan joined forces to tackle this space, they had been contemplating this idea.

Both of them studied at IIT Madras, went to the same exchange program in Singapore, and were colleagues at Kolkata-headquartered conglomerate ITC. While working there, they realized that part of a product manager’s job at the firm was dealing with gazillions of suppliers and the manufacturing items they offered.

The process was archaic: There were no databases, and people couldn’t track shipments.

The early version of Zetwerk, which was a database of suppliers, was a direct response to this. But after listening to requests from customers, the startup saw a bigger opportunity and transformed itself into a full-fledged marketplace with integrations with third-party vendors. Once a firm has placed an order, Zetwerk allows them to keep tabs on the progress of manufacturing and then the shipping. There are also quality checks in place.

Zetwerk operates in such a unique space today — Shailesh Lakhani, managing director at Sequoia India, says the startup has defined a new category of marketplace — that by and large it’s not competing with any other firm in India — or South Asia. (The startup competes with domain project consultants in the offline world.)

The opportunity in India itself is gigantic. According to industry reports, manufacturing today accounts for 14% of India’s GDP. Vaibhav Agarwal, a partner at Lightspeed, estimates that the market is as large as $40 billion to $60 billion in India and global trade-tailwinds that creates opportunity to serve international demand.

As more and more companies expand or shift their manufacturing to India — in part due to import duties imposed by India and geo-political tension with China, the global hub for manufacturing — this opportunity has only grown bigger in recent years.

“India has a lot of depth in manufacturing, but much of it has not been tapped well,” said Acharya.

Zetwerk — which grew 3X last year and reported revenue of $43.9 million in the financial year that ended in March, a 20X growth from the year prior — plans to deploy the new capital to expand to more areas of categories, and broaden its technology stack. Consumer goods (which covers items such as mixer grinders and TVs) is an area Zetwerk expanded to last year, and said it accounts for 15% of the revenue it generated in the last six months.

Currently 25 of its customers are in the U.S., Canada, Europe and other international markets. Acharya said the startup plans to open offices overseas this year as it scouts for more international customers. 

“We are excited to partner with Zetwerk on the next leg of their journey, as they expand their value proposition globally. Zetwerk’s operating system for manufacturing has digitized multiple supply chains end-to-end, ensuring on-time delivery and high quality standards. This has led to rapid growth in India and internationally, with the potential to quickly become one of the most important manufacturing platforms globally,” said Neil Shah, partner at Greenoaks Capital, in a statement.