Kraftful raises $1M to help smart home companies make better apps

If a thousand companies make their own smart light bulb, do a thousand companies also have to design a light switch app to control them?

Kraftful, a company out of Y Combinator’s Summer 2019 class, doesn’t think so. Kraftful builds the myriad components that an IoT/smart home company might need, puzzle piecing them together into apps for each company without requiring them to reinvent the light switch (or the padlock button, or the smart thermostat dial) for the nth time.

Because no company wants an app that looks identical to a competitor’s, much of what Kraftful produces is built to be tailored to each company’s branding — all the surface-level stuff, like iconography, fonts, colors, etc. are all customizable. Under the hood, though, everything is built to be reusable.

This focus on finding the parts that can be built once makes sense, especially given the team’s background. CEO Yana Welinder and CTO Nicky Leach were previously head of Product and a senior engineer, respectively, at IFTTT — the web service made up of a zillion reusable, interlinking “recipe” applets that let you hook just about anything (Gmail, Instagram, your cat’s litter box, whatever) into anything else to let one trigger actions on the other.

Kraftful founders Nicky Leach and Yana Welinder

So why now? More smart devices are coming onto the market every day, many of them from legacy appliance companies that don’t have much (or any) history in building smartphone apps. Good apps are the exception — the Philips Hue app is one of the better ones out there, and even it’s a little wonky sometimes. Many of them are… really bad.

Bad apps get bad App Store reviews, and bad reviews dent sales. And even for those who dive in and buy it without checking the reviews first, bad apps means returned devices. According to this iQor survey from 2018, 22% of smart home customers give up and return the products before getting them to work.

“We kind of looked around and realized that 80% of all smart home apps have zero, one or two stars on the App Store,” Welinder tells me.

Knowing what’s working and what’s not with buyers is a strength of Kraftful’s approach; behind the scenes, they can run all sorts of analytics on how users are actually interacting with components in the apps they’re powering and adjust all of them accordingly. If they make a tweak to the setup process in one app, do more users actually get all the way through it? Great. Now roll that out everywhere.

“If you look at some of the leading smart lock apps, they all have very… very similar interfaces. They’ve basically gotten to a standardized user experience, but they’ve all be developed individually,” says Welinder. “So all of these companies are spending the resources designing and developing these apps, but they’re not getting the benefit of being standardized across the board and being able to leverage data from all of these apps to be able to improve them all at once”

Kraftful builds the app for both iOS and Android, tailors it to the brand’s needs, offers cloud functionality like push notifications and activity history, provides analytics for insights on how users are actually using an app and keeps everything working as OS updates roll out and as device display sizes grow ever larger.

Of course, the entire concept of a dedicated app for a smart home device has some pretty fierce competition — between Apple’s HomeKit and Google Home, the platform makers themselves seem pretty set on gobbling up much of the functionality. But most buyers still expect their shiny devices to have their own apps — something branded and purpose-built, something for the manual to point them to. Power users, meanwhile, will always want to do things beyond what the all-encompassing solutions like HomeKit/Home are built for.

Folks at Google seem to agree with Kraftful’s approach — the team counts the Google Assistant Investments Program as one of the investors in the $1 million they’ve raised. Other investors include YC, F7 Ventures, Cleo Capital, Julia Collins (co-founder of Zume Pizza and Planet Forward), Lukas Biewald (co-founder of CrowdFlower), Nicolas Pinto (co-founder of Perceptio) and a number of other angel investors.

Welinder tells me they’re already working with multiple companies to start powering their apps; NDAs prevent her from saying who, at this point, but she notes that they’re “some of the largest brands that provide smart lights, plugs/switches, thermostats and other smart home products.”

Here are all 21 companies from Alchemist Accelerator’s latest batch

We’re down in Sunnyvale, CA today, where Alchemist Accelerator is hosting a demo day for its most recent batch of companies. This is the 23rd class to graduate from Alchemist, with notable alums including LaunchDarkly, MightyHive, Matternet, and Rigetti Computing. As an enterprise accelerator, Alchemist focuses on companies that make their money from other businesses, rather than consumers.

21 companies presented in all, each getting five minutes to explain their mission to a room full of investors, media, and other founders.

Here are our notes on all 21 companies, in the order in which they presented:

i-50: Uses AI to monitor human actions on production lines, using computer vision to look for errors or abnormalities along the way. Founder Albert Kao says that 68% of manufacturing issues are caused by human error. The company currently has 3 paid pilots, totalling $190k in contracts.

Perimeter: A data visualization platform for firefighters and other first responders, allowing them to more quickly input and share information (such as how a fire is spreading) with each other and the public. Projecting $1.7M in revenue within 18 months.

Einsite: Computer vision-based analytics for mining and construction. Sensors and cameras are mounted on heavy machines (like dump trucks and excavators). Footage is analyzed in the cloud, with the data ultimately presented to job site managers to help monitor progress and identify issues. Founder Anirudh Reddy says the company will have $1.2M in bookings and be up and running on 2100 machines this year.

Mall IQ: A location-based marketing/analytics SDK for retail stores and malls to tie into their apps. Co-founder Batu Sat says they’ve built an “accurate and scalable” method of determining a customer’s indoor position without GPS or additional hardware like Bluetooth beacons.

Ipsum Analytics: Machine learning system meant to predict the outcome of a company’s ongoing legal cases by analyzing the relevant historical cases of a given jurisdiction, judge, etc. First target customer is hedge funds, helping them project how legal outcomes will impact the market.

Vincere Health: Works with insurance companies to pay people to stop smoking. They’ve built an app with companion breathalyzer hardware; each time a user checks in with the breathalyzer to prove they’re smoking less, the user gets paid. They’ve raised $400k so far.

Harmonize: A chat bot system for automating HR tasks, built to work with existing platforms like Slack and Microsoft Teams. An employee could, for example, message the bot to request time off — the request is automatically forwarded to their manager, presenting them with one-click approve/deny buttons which handle everything behind the scenes. The company says it currently has 400 paying customers and is seeing $500k in ARR, projecting $2M ARR in 2020.

Coreshell Technologies: Working on a coating for lithium-Ion batteries which the company says makes them 25% cheaper and 50% faster to produce. The company’s co-founder says they have 11 patents filed, with 2 paid agreements signed and 12 more in the pipeline.

in3D: An SDK for 3D body scanning via smartphone, meant to help apps do things like gather body measurements for custom clothing, allow for virtual clothing try-ons, or create accurate digital avatars for games.

Domatic: “Intelligent power” for new building construction. Pushes both data and low-voltage power over a single “Class 2” wire , making it easier/cheaper for builders to make a building “smart”. Co-founder Jim Baldwin helped build Firewire at Apple, and co-founder Gladys Wong was previously a hardware engineer at Cisco.

MeToo Kit: a kit meant to allow victims of sexual assault or rape to gather evidence through an at-home, self-administered process. Co-founder Madison Campbell says that they’ve seen 100k kits ordered by universities, corporations, non-profits, and military organizations. The company garnered significant controversy in September of 2019 after multiple states issued cease-and-desist letters, with Michigan’s Attorney General arguing that such a kit would not be admissible in court. Campbell told Buzzfeed last year that she would “never stop fighting” for the concept.

AiChemist Metal: Building a thin, lightweight battery made of copper and cellulose “nanofibers”. Co-founder Sergey Lopatin says the company’s solution is 2-3x lighter, stronger, and cheaper than alternatives, and that the company is projecting profitability in 2021. Focusing first on batteries for robotics, flexible displays, and electric vehicles.

Delightree: A task management system for franchises, meant to help owners create and audit to-dos across locations. Monitors online customer reviews, automatically generating potential tasks accordingly. In pilot tests with 3 brands with 16 brands on a waitlist, which the company says translates to about $400k in potential ARR.

DigiFabster: A ML-powered “smart quoting” tool for manufacturing shops doing things like CNC machining to make custom parts and components. Currently working with 125 customers, they’re seeing $500k in ARR.

NachoNacho: Helps small/medium businesses monitor and manage software subscriptions their employees sign up for. Issues virtual credit cards which small businesses use to sign up for services; you can place budgets on each card, cancel cards, and quickly determine where your money is going. Launched 9 months ago, NachoNacho says it’s currently working with over 1600 businesses.

Zapiens: a virtual assistant-style tool for sharing knowledge within a company, tied into tools like Slack/Salesforce/Microsoft 365. Answers employee questions, or uses its understanding of each employee’s expertise to find someone within the company who can answer the question.

Onebrief: A tool aiming to make military planning more efficient. Co-founder/Army officer Grant Demaree says that much of the military’s planning is buried in Word/Powerpoint documents, with inefficiencies leading to ballooning team sizes. By modernizing the planning approach with a focus on visualization, automation and data re-usability, he says planning teams could be smaller yet more agile.

Perceive: Spatial analytics for retail stores. Builds a sensor that hooks into existing in-store lighting wiring to create a 3D map of stores, analyzing customer movement/behavior (without face recognition or WiFi/beacon tracking) to identify weak spots in store layout or staffing.

Acoustic Wells: IoT devices for monitoring and controlling production from oil fields. Analyzes sound from pipes “ten thousand feet underground” to regulate how a machine is running, optimizing production while minimizing waste. Charges monthly fee per oil well. Currently has letters of intent to roll out their solution in over 1,000 wells.

SocialGlass: A marketplace for government procurement. Lets governments buy goods/services valued under $10,000 without going through a bidding process, with SocialGlass guaranteeing they’ve found the cheapest price. Currently working with 50+ suppliers offering 10,000 SKUs.

Applied Particle Technology: Continuous, realtime worker health/safety tracking for industrial environments. Working on wireless, wearable monitors that stream environmental data to identify potential exposure risks. Focusing first on mining and metals industries, later moving into construction, firefighting, and utilities environments.

Microsoft releases tools to let devs start building for its upcoming dual-screen devices

 

As we learned back in October, Microsoft has been cracking away at not one, but two dual screen devices: Surface Duo, and Surface Neo. Surface Duo will run Android, while the Surface Neo will run on a special fork of Windows 10 dubbed “Windows 10 X”.

This morning the company is pulling back the curtain a bit, debuting its first batch of dual-screen developer tools and shedding some light on how apps can utilize that second screen.

While a developer kit for the Android-powered Duo has been made available immediately, the company says dev tools for the Windows-powered Neo will arrive in “the coming weeks”, with a target date of February 11th.

By default, says Microsoft, apps on these dual-screen devices will only occupy one screen. Users can elect to “span” the app to make it stretch across both — but, at least for now, it’s not something an app can force to happen.

While simply stretching an app to fill both screens is one approach, Microsoft offered up a few alternative “pattern ideas” to better utilize the form factor:

Meanwhile, Microsoft is also starting to build out web standards for dual-screen devices — APIs for developers to easily detect dual-screen devices, for example, allowing them to adapt their web apps accordingly. The company says preview builds of Microsoft Edge with early dual-screen APIs should start shipping “soon”.

While no specific launch date has been given for either device, Microsoft has given both a launch window of sometime around the Holidays of 2020. Getting these dev tools out sooner than later, then, makes sense — while it might look neat, two screens aren’t inherently better than one. For the concept to ever take off, Microsoft needs developers to find the novel ways to use that second screen; the ways in which having a pair of screens really makes things better, rather than just… different.

LaunchDarkly CEO Edith Harbaugh explains why her company raised another $54M

This week, LaunchDarkly announced that it has raised another $54 million. Led by Bessemer Venture Partners and backed by the company’s existing investors, it brings the company’s total funding up to $130 million.

For the unfamiliar, LaunchDarkly builds a platform that allows companies to easily roll out new features to only certain customers, providing a dashboard for things like “canary launches” (pushing new stuff to a small group of users to make sure nothing breaks) or launching a feature only in select countries or territories. By productizing an increasingly popular development concept (“feature flagging”) and making it easier to toggle new stuff across different platforms and languages, the company is quickly finding customers in companies that would rather not spend time rolling their own solutions.

I spoke with CEO and co-founder Edith Harbaugh, who filled me in on where the idea for LaunchDarkly came from, how their product is being embraced by product managers and marketing teams and the company’s plans to expand with offices around the world. Here’s our chat, edited lightly for brevity and clarity.

Formlabs CEO on the state of 3D printing and its remaining challenges

3D printing isn’t the buzzy, hype-tastic topic it was just a few years ago — at least not with consumers. 3D printing news out of CES last week seemed considerably quieter than years prior; the physical booths for many 3D printing companies I saw took up fractions of the footprints they did just last year. Tapered, it seems, are the dreams of a 3D printer in every home.

In professional production environments, however, 3D printing remains a crucial tool. Companies big and small tap 3D printing to design and test new concepts, creating one-off prototypes in-house at a fraction of the cost and time compared to going back-and-forth with a factory. Sneaker companies are using it to create new types of shoe soles from experimental materials. Dentists are using it to create things like dentures and bridges in-office, in hours rather than days.

One of the companies that has long focused on pushing 3D printing into production is Formlabs, the Massachusetts-based team behind the aptly named Form series of pro-grade desktop 3D printers. The company launched its first product in 2012 after raising nearly $3 million on Kickstarter; by 2018, it was raising millions at a valuation of over a billion dollars.

Hipmunk’s co-founders tried to buy it back before the shutdown

In a little over a week, the travel metasearch engine Hipmunk will shut down. It’s an abrupt and disappointing end for a product acquired by SAP Concur around three and a half years ago.

It didn’t have to go this way, though. We’re now hearing that Hipmunk co-founders Adam Goldstein and Reddit CEO Steve Huffman (both of whom left Hipmunk in years prior) made an offer to buy the company back rather than have it shut down, to no avail.

As first reported by Skift and confirmed with our own sources, Hipmunk employees at an all-hands meeting today asked about the buyback offer and were told it wasn’t going to happen.

Why SAP Concur would opt to shut down Hipmunk rather than spin it back out isn’t entirely clear, though these things are undoubtedly more complicated than the parent company saying “Oh, you want it back? Sure! Take it!”

A few years post-acquisition, Hipmunk DNA has intertwined with Concur’s to some degree — including a product for small-to-medium businesses straight up called “Concur Hipmunk” — so there’s presumably interest on SAP Concur’s end in keeping engineers who know all the relevant codebases around. Hipmunk also had patents that switched hands in the acquisition (and more were approved post-acquisition), and it’s possible SAP doesn’t want to let those go. It’s also entirely feasible that the offer came too late, or just wasn’t significant enough, for a company as large as SAP to slam the brakes and start detangling things.

A rep for SAP Concur declined to comment.

Four years after being acquired, Hipmunk is shutting down

If you’re still using Hipmunk for trip planning, it’s time to find another site. Sooner than later, really.

The Hipmunk team announced this afternoon that it’ll be shutting down in just about a week’s time. Come January 23rd, says the blog post, both the Hipmunk website and app will be shuttered.

The team also tweeted:

This shutdown comes right around three and a half years after corporate travel/expensing platform Concur acquired Hipmunk back in September of 2016.

Founded by Adam J. Goldstein and Reddit co-founder Steve “spez” Huffman, Hipmunk was one of the first well-made “metasearch” travel sites. It scrounged up flights (and hotels/car rentals/etc.) from across myriad services like Expedia, Priceline, etc., presenting all the times/prices in one big skimmable interface.

The team had an upbeat tone in the original post announcing the acquisition, with Goldstein noting that “Hipmunk isn’t going anywhere.” Alas, things change in four years. Goldstein parted ways with the team at the end of 2018.

In an FAQ about the announcement, Hipmunk clarifies that all accounts will go dark come January 23rd — and while it says existing reservations won’t be impacted by this, they note that all booking matters should go through whichever final provider they pointed you to.

Pokémon Sword and Shield are getting downloadable expansions this year

If you’ve already completed your Pokédex and battled your way up to become the Champion of Galar, you might’ve figured you were just about done with Pokémon Sword/Shield.

Surprise! More is on the way — for a price, that is.

This morning Nintendo announced that it’s working on not one, but two downloadable expansions for Sword and Shield: The Isle of Armor (arriving June 2020) and The Crown Tundra (coming sometime “in the fall of 2020”). A $30 “Expansion Pass” that’s up for pre-order today gets you both downloads when they arrive.

While this definitely isn’t the first time Nintendo has dabbled with DLC, it is the first time they’ve done so with a main series Pokémon title. Considering that Sword/Shield is one of the best-selling Switch games of all time, it makes sense that Nintendo isn’t quite ready to be done with it.

So what’s new in the expansions?

Each pack will bring new areas to explore, along with new characters, storylines and, of course, Pokémon. Nintendo and GameFreak aren’t getting too specific about how many new Pokémon we’ll see, but do note that the expansions will include brand new monsters, Gigantamax and Galarian forms of existing ‘mon, plus support for an unspecified number of past Pokémon transferable from previous Pokémon titles.

While the expansions won’t ship for a few months, the companies are releasing what they call “a small slice” of the new stuff today via a free update.

Here’s everything Google announced at CES 2020

Another year, another blast of Google Assistant news on the first official day of CES.

Google slimmed things down a touch for CES this year, though they’ve still got a big presence here. While they didn’t build a whole damn amusement park ride this time around, they’ve still got a massive two-story booth (complete with slides?) parked right outside the front doors of the Las Vegas Convention Center.

As with last year, just about everything Google is showing off at CES 2020 is focused around the company’s voice-powered AI helper, Google Assistant.

Here’s what’s new:

  • Webpage reading: Been meaning to read that long article all day, but don’t have 20 minutes to stare at your phone? Folks on Android devices will soon be able to say “Hey Google, read this page” and Assistant will fire up its neural networks to generate a pretty lifelike reading, with the system only reading the relevant text (assuming all is working as planned) while avoiding mentioning things like social sharing buttons or the page’s myriad navigation options. Google says it also wants to make the page autoscroll/highlight text as it reads, though it sounds like that may come a bit later down the road.
  • Scheduled actions: You’ll soon be able to make one-off requests for things you want to happen later in the day, like “Hey Google, turn on the lights at 6 pm.” Sorta surprising this wasn’t already a thing.
  • Sticky notes: Ever written something on a Post-it and put it wherever someone would see it as soon as they walked in the door? Google is taking that idea and putting it on their smart displays, allowing you to say things like “Hey Google, leave a note that says ‘Don’t forget to pack’ ” to pin a note to the lock screen (visible by all) accordingly.

  • Speed Dial: If that sticky notes feature is meant to replace the front door Post-it, this one is meant to replace that list of important phone numbers stuck to the fridge door for the babysitter. You’ll be able to add a handful of phone numbers to a smart display’s lock screen, allowing anyone to quickly call those contacts with a tap or voice command.
  • More businesses using Interpreter Mode: Announced at CES last year, interpreter mode allows an Assistant-powered smart display to translate a conversation between two people, each speaking a different language. Google says more businesses have committed to using it this year, including American Airlines, HSBC banks and a handful of hotels around Vegas, San Francisco, LA, Japan and Qatar.
  • Uh, forget I said that: Google Assistant isn’t supposed to record anything you say unless you start the sentence with “Hey Google”… but, well, that doesn’t always work. Sometimes things on TV will cause Assistant to perk up its ears; other times you might be mid-conversation and only realize you somehow caught Assistant’s attention when it responds “Sorry, I can’t help with that.” With that in mind, you’ll now be able to say “Hey Google, that wasn’t for you” to have it wipe its history of the last thing you said.

The one catch: As with most new Google Assistant features, the company isn’t getting too specific about when this stuff is rolling out, saying only that it’ll come “later this year.”

CES 2020 coverage - TechCrunch

Adobe CTO says AI will ‘democratize’ creative tools

Adobe CTO Abhay Parasnis sees a shift happening.

A shift in how people share content and who wants to use creative tools. A shift in how users expect these tools to work — especially how much time they take to learn and how quickly they get things done.

I spoke with Parasnis in December to learn more about where Adobe’s products are going and how they’ll get there — even if it means rethinking how it all works today.

“What could we build that makes today’s Photoshop, or today’s Premiere, or today’s Illustrator look irrelevant five years from now?” he asked.

In many cases, that means a lot more artificial intelligence; AI to flatten the learning curve, allowing the user to command apps like Photoshop not only by digging through menus, but by literally telling Photoshop what they want done (as in, with their voice). AI to better understand what the user is doing, helping to eliminate mundane or repetitive tasks. AI to, as Parasnis puts it, “democratize” Adobe’s products.

We’ve seen some hints of this already. Back in November, Adobe announced Photoshop Camera, a free iOS/Android app that repurposes the Photoshop engine into a lightweight but AI-heavy interface that allows for fancy filters and complex effects with minimal effort or learning required of the user. I see it as Adobe’s way of acknowledging (flexing on?) the Snapchats and Instas of the world, saying “oh, don’t worry, we can do that too.”

But the efforts to let AI do more and more of the heavy lifting won’t stop with free apps.

“We think AI has the potential to dramatically reduce the learning curve and make people productive — not at the edges, but 10x, 100x improvement in productivity,” said Parasnis.

“The last decade or two decades of creativity were limited to professionals, people who really were high-end animators, high-end designers… why isn’t it for every student or every consumer that has a story to tell? They shouldn’t be locked out of these powerful tools only because they’re either costly, or they are more complex to learn. We can democratize that by simplifying the workflow.”