TechStars, a Boulder-based incubator, demoed their products in Mountain View, Calif. and did a good job of building on the city's reputation as a budding center for high-tech. Almost down to a team, each of companies that presented this morning felt strong.
The TechStars demo reviews:
Devver. The company's mission is to radically improve developer productivity. Aside from wasting time on social networks and video games, software engineers who are testing their code spend an inordinate amount of time waiting as the test runs. The problem is that every test resides in single units on individual desktops. Devver, on the other hand operates in the cloud and spreads the job across multiple servers. As a result, the whole thing happens 75 percent faster, every time. The service, which currently only works to test applications that use Ruby on Rails, is $100 per month per user. The advantage of Devver's model is that the user need not pay infrastructure and maintenance costs. The disadvantage is a company running its own hardware has sole access to its code -- control that vanishes when it uploads its entire codebase to a third party. Also, Devver is built on Amazon's S3 service, so the quality of the security of the code rests primarily on Amazon. You make the call.
Occipital, however, it's clear to me that those other two products' time has passed. Occipital uses machine vision to identify nearly any object in an image. Automatically identifying and labeling landmarks is one thing. Occiptial does it, but so does a 3D mapping company called Everyscape. What Everyscape can't do is recognize everything else. Apparently, Occipital can. If you, say, travel with a favorite stuffed animal and photograph it everywhere you go, Occipital will recognize it, label it and let you find every picture that contains it later on. If multiple people upload multiple photographs from the same event around the same time, Occiptial will figure out that an event just happened and classify the photographs accordingly. Doing this right is really, really hard, yet with two people, Occipital seems to have done it. This team is scary good.
Gyminee combines in-depth excerise and nutrition guides with social networking and "biggest loser" competitions to try to help people achieve their weight-loss goals. Targeted at affluent male gym goers with big goals, Gyminee gives users tons of metrics, videotaped exercise demonstrations and the ability to record their progress on their iPhones and get encouraging feedback from friends and fellow users. It sits in a similar space as SparkPeople and DietTV. While the former seems to be riding high, the latter seems to have lost the majority of its users after four months. Much of Gyminee's service is free, but users will have to pay $15 every three months to get features like meal planning and personal training guides. If the company is smart, it will use Facebook's FriendConnect to help register users, instead of making users build yet another new profile.
BrightKite, which has been around for a while, is a location-based social network, and today the company demoed its forthcoming iPhone app. It's a bit behind Loopt and Whrrl in getting itself onto the iPhone, but from the looks of it, it could displace them. BrightKite's app seems to offer a wider range of features than its competitors, including location bookmarking, the ability to find and learn about fellow users in the same area, and "place snapping," which uses your location history to make educated guesses about the places you're standing. I'll do a more in-depth review when I test the app out myself.
TravelFli wants to help the 17 million "elite" frenquent fliers who fly an average of nine trips in a year track and exploit the available rewards. From one web-based interface, these elites can manage their frequent flier miles, credit card points and rental car and hotel loyalty programs. Currently, they cannot use TravelFli to actually book anything, but the company says this functionality is coming. The most striking thing about these presenters -- aside from the to-die-for pilot's outfits they wore -- was their claim that each TravelFli user is worth about $135 per year. TravelFli is not the kind of startup that gets excitable tech journalists to leap out of their seats, but at least it thinks it has a way to make cash. The company has raised $500K so far, but interest has been high enough to make it consider going after a bigger round.
IntenseDebate came onstage, talked for about three minutes and then announced it had just been bought. Automattic, makers of Wordpress, was the buyer. You can read more details about that in our earlier post.
Ignighter's take on that problem is to try to create a social network for groups, based around the idea of meeting people you'd want to date within a group that your group is friends with. You and your friends create a group on the site, add information about each of yourselves, and message people in other groups. The company also lets you share what your plans are, so you can actually go on real-life dates. It also offers an iPhone application. Rival startups include MIXTT, another group-themed dating site that launched at TechCrunch50. Ignighter, meanwhile, launched last month and it says it has more than 10,000 total users so far. It hopes to make money through a variety of revenue streams, including sponsored dates, virtual goods that you can buy for the hot ones, and ads.
Foodzie.The company has taken the "buy hand-made" ethos of Etsy and applied it to food. Instead of warehousing the goods and putting an expensive layer between seller and buyer, like Amazon Grocery and iGourmet, Foodzie's site connects buyers directly to producers. It takes a $0.60 cent transaction fee and 20 percent commission on all sales, which is much less than the 50 percent typically taken at retail. Right now, the site has only a handful of producers but has raised a seed round to expand. This concept is easy to copy, so the relationships with the producers -- and the ability of the producers to scale with demand -- will make or break the company.
Artist Exploder comes in. The company has built a tool that lets musicians build their own branded Facebook application in around five minutes. Once an app is in place, the artist can push music, merchandise and targeted messages right to their listeners. The goal is to capture the die hards -- the ones that spring for every song that comes out and buy t-shirts at ludicrous prices. iLike is already established in this game, and startups like TopSpin Media and LP33 (formerly MyAWOL) have a similar vision. It'll be interesting to see who makes the big play in this game. As a side note, before they started Artist Exploder, the founders had a company called MadKast. As we reported earlier, the content-sharing widget company, ShareThis, just bought it.
EventVue might be able to help. The company, whose product was used by the organizers of recent DEMOfall '08 tech conference, takes the list of registered attendees, creates mini-profiles for each one and crawls the web to fill them out with as much data as possible. Once the database is populated, attendees can find their profiles and fill in the blanks, writing in their interests and industry specialties and adding tags to describe both and make themselves easy to find. Since professional networking is one of primary benefits of going to conferences, a tool that significantly facilitates the process is undoubtably useful. EventVue's founders claim that as much as 43 percent of DEMO attendees used EventVue for an average of one hour each.
As a quick note for engineers interested in the Boulder tech scene: The growing number of startups (around 50, by one count) has apparently been draining the talent pool. Now some of the companies have banded together to offer a free week-long tour to check out the local companies. It's called Boulder.me, and you can also read more about it on the Gnip company blog, one of the tour sponsors.
Eric Eldon contributed writing and reporting to this story.