If you’re looking to hire a bad ass programmer, fielding resumes can start to feel like an exercise in futility. They’re good for quickly filtering out folks who are clearly applying to everything — but when everyone in the industry has some crazy made-up senior ninja/rockstar/space cadet title, when everyone considers themselves a coder, and when “Proficient with C” can mean two entirely different things based on a person’s ego, that’s about all they’re good for.
Hackermeter, part of the most recent Y-combinator class, thinks they have a better alternative: a coder score.
Hackermeter is based around the concept of coding challenges. The better you perform on each challenge, the higher your score. The higher your score, the more enticing you’ll be to a potential employer.
Why do law firms spend, collectively, billions of dollars on commercial legal research databases, when what they are looking up is law — which is in the public domain? How are these databases able to erect these enormously profitable paywalls? The answer is that they provide more than just the raw text of the law. They provide search tools and additional, value-added content on top of the law itself. The two legal research titans, Lexis and Westlaw, employ lawyers to read cases and other legal materials, categorize them, add commentary, and link them together. These services have legitimate value because they all save lawyers time, and time is money — especially in a profession that largely bills its clients in six-minute increments. That’s why these expensive tools exist, even in the Internet age. As one lawyer put it, after trying to get by on only free legal research tools, he tried Westlaw and was an immediate convert who now happily pays for the service.
Two young lawyers thinks they can disrupt the legal research giants by applying the lessons of Wikipedia and crowdsourcing their own comparable set of annotated law.
Joanna Huey attended Harvard Law School, where she was president of the Harvard Law Review, and Jacob Heller attended Stanford Law School, where he was president of the Stanford Law Review. They later served together as clerks for Judge Michael Boudin at the U.S. Court of Appeals for the First Circuit in Boston, and worked at law firms. Both were dissatisfied with the available research tools and their hefty price tags, which put the poor at a competitive disadvantage in the justice system.
Unlike many lawyers, neither Huey nor Heller are afraid of technology, so they decided to do something about it. Huey’s undergraduate degree is in physics, and Heller was a web developer before law school. They applied to Y Combinator, and were accepted. They’re now emerging from the program and ready to launch their company: Casetext.
While traditional analytics detail what users tap in an app, they don’t show what happens in between. If only developers could watch their users in action. With Y Combinator-backed Watchsend, developers can get a peek by remotely recording users’ screens, and learn how to refine their apps.
Whether the app is in beta or live, Watchsend users can specify exactly what pages they want to record, which are then encrypted and secured in its data base. All recordings come with screen annotations of taps, swipes and presses, as well as a timeline of the user’s actions. App developers can then go in and view them, using very specific search terms to navigate the vast number of videos collected.
Watchsend is meant to complement traditional app testing methods such as conversion rates and A/B testing, not replace them. ”There’s a lot of stuff that you get from traditional analytics that’s absolutely indispensable, and we don’t claim to be able to give you any of that,” Gulrajani says. “But there’s a lot of stuff that analytics misses out on, stuff that you didn’t think to track.”
Instead, it’s offering an alternative solution for apps where user progress is harder to track. For example, analyzing data from mobile gaming apps can show at what level users are dropping off, but its a lot more difficult to determine exactly what the user is doing wrong. Maybe the ladder to the next level is blocked by in-game shrubbery, or that last gem to be collected is too hard to find. Analytics also show which pages are reached by users, but not how far down they scroll before clicking “next” or “buy”.
Over a year after the announcement Of Google Glass, many folks I talk to still seem to be misunderstanding what Glass can actually do.
“It’ll be great for Augmented Reality!” they say, assuming that Glass can render objects directly into your full view of the world (it can’t.) “Ooh! It’ll be like Minority Report!”, expecting Glass’ camera to pick up your every hand wave (it doesn’t.)
Then they try on a pair and realize that… well, that’s not what Glass is. But it’s what Meta is aiming to be — and their first (read: still a bit rough) version is going on sale to the public starting today.
To picture the Meta, picture a pair of glasses — or, more accurately in its current stage, a pair of safety goggles. Put a translucent, reflective surface in each eye piece, displaying images on top of your field of view as piped out of a tiny projector built into each arm of the frames. Take a couple tiny RGB/Infrared cameras — essentially a miniature Kinect — and strap them to the frame. That’s the Meta.
The Meta then plugs into another device to help it with the data crunching; right now, that’s a laptop. Moving forward, it’ll be your phone.
After flying under the radar for a bit over a year, Meta debuted itself to the world on Kickstarter back in May. By the end of their campaign, they’d nearly doubled their original goal of $100,000. They promised to ship those units to their backers by the end of this month, and they say they’re on track to meet that deadline — so now they’re opening up pre-sales of the next iteration to everyone.
To be clear, the hardware they’re launching today is still quite early. It’s perhaps a bit past the “Developers Only” level, but it’s still mostly meant for the hardcore early-adopters and tinkerers. Hell, its early state is reflected in its very name; this model is called the META.01, suggesting many a revision to come. The META.01 units are going up for sale at $667, with plans to begin shipping in November.
Also watch the behind-the-scenes review by Reddit (YC S05) cofounder Steve Huffman as he uses the META.01 prototype...
In the past few years, it seems like there’s been the explosion of startups trying to convince local businesses to sign up for their marketing or loyalty tools, especially in the San Francisco Bay Area. In fact, LocalOn co-founder Shahbano Imran recalled going door-to-door trying to convince businesses to sign up and discovering that “small businesses are getting pitched by 20 startups a day.”
As a result, she said LocalOn’s initial efforts were “a complete failure” because “nobody wanted to hear from us.” Then she and her co-founder David Tolloupov came up with a better way to reach those businesses — working with local newspapers and merchant associations.
So instead of getting pitched by a random startup, businesses are offered a set of white labeled tools from a publication or an organization that they already trust. (The revenue is split between LocalOn and its resale partner.)
The approach seems to be working for the startup, which is part of incubator Y Combinator’s current class of companies. It has already partnered with 40 merchant associations and two newspapers in the Bay Area — apparently the partnership has generated $50,000 in new business for the East Bay Express newspaper over the past six months, and the OaklandGrown merchant association has seen a 20 percent increase in revenue from annual memberships.
Siasto has received some more support to build out its task management service with $750,000 from One Asia Investment Partners, Y Combinator, Start Fund and other angel investors.
Siasto has developed a robust platform that TechCrunch’s Eric Eldon wrote about in detail last year. Last November, the company added a news feed to the service. With the new funding, Siasto will now extend its development to focus more on its mobile collaboration service. The funding will also be used to expand its service into Asia.
Co-founder Niccolo Pantucci says its mobile strategy is to build an app store in Siasto that features distinct apps that take from Siasto but that have specific functions. There might be linking between different components but each will act as a separate app.
New hardware startup and Y Combinator Summer 2013 cohort member Senic is launching pre-orders for its first product today, a laser rangefinder like the ones sold in hardware stores around the world and used every day by contractors and DIY enthusiasts. The difference is that Senic’s device uses Bluetooth Low Energy to communicate with iOS and Android devices, recording measurements instantly and syncing them to the cloud.
The use of a cloud-based platform for collecting and storing measurement data, as well as Senic’s plans to expand their hardware catalog to include a number of different hardware devices, including micrometers, gauges, and other tools that builders and engineers use regularly in their work. Senic co-founder Toby Eichenwald explained that he got the idea working for his father’s company in Korea, and learned from its customers that the measurement industry was essentially “stuck in the 80s.”
Streak has embedded Gmail into its new iOS app in a novel way to create a service that is turning the 600-million-user platform into a powerful CRM environment.
Nathan Wenzel and Patrik Outericky had a successful services business called Edge Solutions that helped enterprises — especially insurance companies with large portfolios of cases — sort out their legal bills. Although the business wasn’t sexy, it was profitable. But Wenzel and Outericky decided to wind that company down and go into Y Combinator to turn that review process into a scalable product instead. They are now emerging from the accelerator with a new company,SimpleLegal.
SimpleLegal takes as much friction out of the bill review process as possible. All a customer has to do is ask their law firm to copy SimpleLegal on each invoice, and then the magic starts to happen. SimpleLegal’s system ingests the invoice and parses each line item into its database. Natural language processing systems figure out who billed what and for how long — and then that data is run through a machine learning system that flags outliers. One example: the system flagged a line item where a professional billed a half hour for mailing. That might not be too unusual but for the fact that the system knew the thing being mailed was a one-page form. That’s pretty smart.