Connect with us


People are fed up with broken vaccine appointment tools — so they’re building their own



Across the US, people are clamoring to get the hottest ticket of the season: an appointment for a covid-19 vaccination. The recommended method is to visit a local hospital website or call a hotline. But the results can be frustrating. Families have spent hours on the hotlines, scoured appointments on platforms like the ticket sales site Eventbrite, and posted desperately on local news sites like Patch and Nextdoor. 

The nation’s vaccine rollout has been messy, in other words. Now, desperate people are taking matters into their own hands.

Take Carri Carver. She had spent hours trying to find her father an appointment, once he had qualified for a vaccine under Texas rules. He went to several local pharmacies that were listed on the state’s official site, none of which had any vaccines left. Some told him to “come back later and ask again.”

“I was like, ‘That’s a terrible answer,’” Carver says. “This is going to be a challenge for him. He’s very healthy and in the younger part of the age group. What about the people who are older or can’t go around to check [for vaccines]?”

That very day, January 2, Carver worked from 3 in the afternoon until 11:30 at night to create Covid19 Vaccine TX, a site listing possible vaccination locations across the state. As a digital product designer, she knew that a site like this would have to be easy to read, intuitive to navigate, and quick to update. The idea was that people could upload information about vaccination sites, with each entry answering three questions: Was the vaccine available that day? Was the location taking appointments? Was there a wait list?

Carver loaded the project on the cloud-based spreadsheet service Airtable, posted a link on Reddit, and went to bed. When she woke up the next morning at 7 a.m., one entry was filled out. “At least somebody cares,” she remembers thinking. She spent the rest of the day manually inserting information for about 1,400 locations in the state. “I’ve been going nonstop since,” she says, estimating that she puts in about 40 hours of her free time every week to maintain the site. It has received 50,000 total visitors since launch.

Carver’s site is just one of many solutions created to help people book vaccine appointments—a volunteer effort to fill in the gaps in America’s creaking system.

Another is Vaccinate CA, a site featuring a map of the state. Users can click on their location to see a pop-up that provides the latest information about eligibility requirements, vaccine availability, and contact information so they can book an appointment. Zoelle Egner is one of the organizers. “I wouldn’t actually call it crowdsourcing,” says Egner. “Although we ask the public to let us know if anything we’ve published has gone out of date, we get the information on our website by directly calling vaccination locations and asking medical professionals if the vaccine is available, and how to get it.”

These efforts are rapidly multiplying across the country. Carver’s Texas-based site is working with Oracle to create similar sites in at least 20 states by early next week.

Meanwhile, Leon Wu, a software engineer in New Jersey, heard about VaccinateCA on Twitter and created Vaccinate NJ. Like Carver, Wu created the site in one evening, manually adding information until he could get volunteers to help him verify that everything they’ve posted is correct. “Our tool wouldn’t be very useful if the information can’t be trusted to be accurate,” he says.

Some of these efforts are simply public Google Docs that any user can suggest additions to, which are then vetted. Krystal Knapp created such a spreadsheet of vaccination efforts in Princeton, New Jersey, where she lives, calling it Planet Princeton. Knapp says she started the spreadsheet by “[looking] up all the initial information and … updating it.”

Dan Benamy, a software engineer in Brooklyn, created the site NYC Vaccine List after he tried to use New York’s vaccine finder to get appointments for his grandparents and couldn’t find information on whether locations had availability. With the help of some friends, Benamy built an automated system that checked 45 official vaccination websites every few minutes.

Crowdsourcing information and recruiting volunteers might help reduce the workload of the people behind the sites, but nearly everyone I spoke to was exhausted, balancing the efforts with their day jobs. “It’s pretty much been from when I wake up to when I sleep, minus a few breaks and meals,” Wu says.

Shikha Jain and Halleh Akbarnia, two physicians dealing directly with coronavirus patients in the Chicago area, know a little bit about that. “We have full-time jobs. We do this on nights, weekends, full time,” says Akbarnia of the efforts they’ve embarked on together to get information to Illinois’s most vulnerable residents. “We aren’t getting paid for it.” But physicians feel some guilt about being among the first to be vaccinated, she says, and witnessing the death of her patients and colleagues from the virus made her itch to do something to help.

Akbarnia created a crowdsourced Google Doc listing locations that are accepting appointments. Jain, the COO of the covid-19 advocacy group Impact, has also created a vaccine clearinghouse for health-care workers who are unaffiliated with a hospital, while Akbarnia has been reaching out to pharmacies to help spread the word about the Google Doc.

“What upset me was that there was a lack of information about where to get vaccinated, tested, or get help,” she says. 

These sites are only plugging gaps, and the results are not foolproof. Google Docs will max out at 100 public viewers, which happens frequently, crashing the sites. And such solutions can be less useful for those without a Wi-Fi connection, or who lack tech fluency. Elderly people often are not as comfortable navigating the internet as they are on the phone, so many sites have relied on volunteers to take calls and help guide vulnerable residents. That service extends to those with disabilities or problems with literacy. Crowdsourcing sites are also trying to be creative in reaching out to people who don’t speak English; Jain said some volunteers are visiting churches to combat vaccine disinformation in Spanish, while Carver says Oracle is translating her material for the state’s Spanish-speaking demographic. 

Coordinating volunteers and spreading the word during a pandemic requires multiple apps. Benamy says he uses a shared Discord server to communicate with volunteers. Many also use Airtable. Nearly every group I spoke to has a Twitter presence, hoping to draw the attention of potential volunteers or someone who may know an elderly person in need.

That everyday people are having to do what should be the government’s job is not lost on these volunteers. But most don’t see any use in complaining. “The reason we created this group is because we saw at the beginning of the pandemic there was no national plan to really combat this virus,” says Jain. “Part of it was because it was a new virus, and part of it was that there was no coordinated effort to address this as a public health effort. We saw this as an extension of our role as public health messengers.”

Egner agrees. “We’re not here to point fingers or blame anyone,” she says. “The system is under an incredible amount of stress, and it’s a really complex problem.” 

For many of these volunteers, helping to create these sites is stoking a sense of community that is very much needed after nearly a year of social isolation. “Up until two weeks ago, I’ve personally felt helpless in the face of the whole situation,” Egner says. “We’ve heard from so many people that our website helped them get vaccinated. That’s something I’ll carry with me forever.” 

Carver’s dad eventually did get the vaccine. He and her mother woke up at 3 a.m., drove 45 minutes to their local hospital to book an appointment, returned home, and then came back for their appointment the same day at 1 p.m.

“What 70-year-old should be driving hours at three in the morning to get a vaccination?” Carver says. That’s something a crowdsourced list can’t answer.

Lyron Foster is a Hawaii based African American Musician, Author, Actor, Blogger, Filmmaker, Philanthropist and Multinational Serial Tech Entrepreneur.

Continue Reading


MealMe raises $900,000 for its food search engine



This morning, a food search engine, announced that it has closed a $900,000 pre-seed round. Palm Drive Capital led the round, with participation from Slow Ventures and CP Ventures.

TechCrunch first became familiar with MealMe when it presented as part of the Techstars Atlanta demo day last October, mentioning it in a roundup of favorite startups from a group of the accelerator’s startup cohorts.

The company’s product allows users to search for food, or a restaurant. It then displays price points from various food-delivery apps for what the user wants to eat and have delivered. And, notably, MealMe allows for in-app checkout, regardless of the selected provider.

The service could boost pricing and delivery-speed transparency amongst the different apps that help folks eat, like DoorDash and Uber Eats. But Mealme didn’t start out looking to build a search engine. Instead it took a few changes in direction to get there.

From social network to search engine

MealMe is an example of a startup whose first idea proved only directionally correct. The company began life as a food-focused social network, co-founder Matthew Bouchner told TechCrunch. That iteration of the service allowed users to view posted food pictures, and then find ordering options for what they saw.

While still operating as a social network, MealMe applied to both Y Combinator and Techstars, but wasn’t accepted at either.

The startup discovered that some of its users were posting food pics simply to get the service to tell them which delivery services would be able to bring them what they wanted. From that learning the company focused on building a food search engine, allowing users to search for restaurants, and then vet various delivery options and prices. That iteration of the product got the company into Techstars Atlanta, eventually leading to the demo day that TechCrunch reviewed.

During its time in Techstars, the company adjusted its model to not merely link to DoorDash and others, but to handle checkout inside of its own application. This captures more gross merchandize value (GMV) inside of MealMe, Bouchner explained in an interview. The capability was rolled out in September of 2020.

Since then the company has seen rapid growth, which it measures at around 20% week-on-week. During TechCrunch’s interview with MealMe, the company said that it had reached a GMV run rate of more than $500,000, and was scaling toward the $1 million mark. In the intervening weeks the company passed the $1 million GMV run-rate threshold.

MealMe was slightly coy on its business model, but it appears to make margin between what it charges users for orders and the total revenue it passes along to food delivery apps.

TechCrunch was curious about platform risk at MealMe; could the company get away with offering price comparison and ordering across multiple third-party delivery services without raising the ire of the companies behind those apps? At the time of our interview, Bouchner said that his company had not seen pushback from the services it sends users to. His company’s goal is to grow quickly, become a useful revenue source for the DoorDashes of the world, and then reach out for some of formal agreement, he explained.

“We continue to be a powerful revenue generator and drive thousands of orders to food delivery services per week,” the co-founder said in a written statement. Certainly MealMe found investors more excited by its growth than concerned about Uber Eats or other apps cutting the startup off from their service.

What first caught my eye about MealMe was the realization of how much I would have used it in my early 20s. Perhaps the company can find enough users like my younger self to help it scale to sufficient size that it can go to the major food ordering companies and demand a cut, not merely avoid being cut off.

Continue Reading


Apple supplier Foxconn reaches tentative agreement to build Fisker’s next electric car



Apple supplier Foxconn Technology Group has reached a tentative agreement with electric vehicle startup-turned-SPAC Fisker to develop and eventually manufacture an EV that will be sold in North America, Europe, China and India.

Fisker and Foxconn said Wednesday that a memorandum of understanding agreement has been signed. Discussions between the two companies will continue with the expectation that a formal partnership agreement will be reached during the second quarter of this year. 

Under the agreement, Foxconn will begin production in the fourth quarter of 2023 with a projected annual volume of more than 250,000 vehicles. The electric vehicle will carry the Fisker brand.

Foxconn Technology Group Chairman Young-way Liu touted the company’s vertically integrated global supply chain and accumulated engineering capabilities, noting that it gives the company two major advantages in the development and manufacturing of the key elements of an EV, which includes the electric motor, electric control module and battery.

That supply chain and ability to scale engineering quickly will be critical for Foxconn if it hopes to meet its production target.

“The collaboration between our firms means that it will only take 24 months to produce the next Fisker vehicle — from research and development to production, reducing half of the traditional time required to bring a new vehicle to market,” Young-way Liu said in a statement.

Fisker said production of the Ocean SUV — its first EV and one that is supposed to be built by contract manufacturer Magna — will begin in the fourth quarter of 2022. The company said it plans to unveil a production-intent prototype of the Ocean later this year.

This is not Foxconn’s first foray into electric vehicle manufacturing.

Foxconn announced in January 2020 that it had formed a joint venture with Fiat Chrysler Automobiles to build electric vehicles in China. Under that agreement, each party will own 50% of the venture to develop and manufacture electric vehicles and engage in an IOV, what Foxconn parent company Hon Hai calls the “internet of vehicles” business.

Last month, Foxconn and Chinese automaker Zhejiang Geely Holding Group agreed to form a joint venture focused on contract manufacturing for automakers, with a specific focus on electrification, connectivity and autonomous driving technology as well as vehicles designed for sharing.

The joint venture between Foxconn and Geely will provide consulting services on whole vehicles, parts, intelligent drive systems and other automotive ecosystem platforms to automakers as well as ridesharing companies. Geely said it will bring its experience in the automotive fields of design, engineering, R&D, intelligent manufacturing, supply chain management and quality control while Foxconn will bring its manufacturing and Information and Communication Technology (ICT) know-how.


Continue Reading


Select Star raises seed to automatically document datasets for data scientists



Back when I was a wee lad with a very security-compromised MySQL installation, I used to answer every web request with multiple “SELECT *” database requests — give me all the data and I’ll figure out what to do with it myself.

Today in a modern, data-intensive org, “SELECT *” will kill you. With petabytes of information, tens of thousands of tables (on the small side!), and millions and perhaps billions of calls flung at the database server, data science teams can no longer just ask for all the data and start working with it immediately.

Big data has led to the rise of data warehouses and data lakes (and apparently data lake houses), infrastructure to make accessing data more robust and easy. There is still a cataloguing and discovery problem though — just because you have all of your data in one place doesn’t mean a data scientist knows what the data represents, who owns it, or what that data might affect in the myriad of web and corporate reporting apps built on top of it.

That’s where Select Star comes in. The startup, which was founded about a year ago in March 2020, is designed to automatically build out metadata within the context of a data warehouse. From there, it offers a full-text search that allows users to quickly find data as well as “heat map” signals in its search results which can quickly pinpoint which columns of a dataset are most used by applications within a company and have the most queries that reference them.

The product is SaaS, and it is designed to allow for quick onboarding by connecting to a customer’s data warehouse or business intelligence (BI) tool.

Select Star’s interface allows data scientists to understand what data they are looking at. Photo via Select Star.

Shinji Kim, the sole founder and CEO, explained that the tool is a solution to a problem she has seen directly in corporate data science teams. She formerly founded Concord Systems, a real-time data processing startup that was acquired by Akamai in 2016. “The part that I noticed is that we now have all the data and we have the ability to compute, but now the next challenge is to know what the data is and how to use it,” she explained.

She said that “tribal knowledge is starting to become more wasteful [in] time and pain in growing companies” and pointed out that large companies like Facebook, Airbnb, Uber, Lyft, Spotify and others have built out their own homebrewed data discovery tools. Her mission for Select Star is to allow any corporation to quickly tap into an easy-to-use platform to solve this problem.

The company raised a $2.5 million seed round led by Bowery Capital with participation from Background Capital and a number of prominent angels including Spencer Kimball, Scott Belsky, Nick Caldwell, Michael Li, Ryan Denehy and TLC Collective.

Data discovery tools have been around in some form for years, with popular companies like Alation having raised tens of millions of VC dollars over the years. Kim sees an opportunity to compete by offering a better onboarding experience and also automating large parts of the workflow that remain manual for many alternative data discovery tools. With many of these tools, “they don’t do the work of connecting and building the relationship,” between data she said, adding that “documentation is still important, but being able to automatically generate [metadata] allows data teams to get value right away.”

Select Star’s team, with CEO and founder Shinji Kim in top row, middle. Photo via Select Star.

In addition to just understanding data, Select Star can help data engineers begin to figure out how to change their databases without leading to cascading errors. The platform can identify how columns are used and how a change to one may affect other applications or even other datasets.

Select Star is coming out of private beta today. The company’s team currently has seven people, and Kim says they are focused on growing the team and making it even easier to onboard users by the end of the year.

Continue Reading