If you’ve ever filled out an online job application, you know they’re often complicated. Often adding to the headache is just getting to application stage to begin with. If the jobs aren’t advertised on major services like Monster.com, then those posted positions can be lost in the so-called “Deep Web” or the “Invisible Web.” But don’t turn around, the Komissarchiks in town (uh-oh, uh-oh).
Touring various company websites everyday is a time consuming chore often accompanied by sifting through channels and forms within the site (i.e., following links from Careers to Location to Field/Job Type).
A father-daughter team in the San Francisco area claims they have developed a way to streamline this process. Julia and Edward Komissarchik’s Glenbrook spiders act more like spies or scouts than crawlers. The artificially intelligent “scouts” work like safe crackers.
“The way to think of it is, you case the joint,” she said. “The scout goes through the form and tries a few options to see what the results will be. Then you have a mastermind or safecracker who gets all this information from the scout and devises a method to open the forms.”
The information gleaned from these sites isn’t private or locked up in any way. It’s generically available information that’s just hard to get to.
Recently, the pair launched a beta called the “Glendor Showcase” for the San Francisco Bay area only that utilizes GoogleMaps to tell searchers more about job locations.
The name “Glendor” came about as a contraction of “gleaning d’or,” a reference to searching for gold in 49er fashion.
“Our claim to fame is to be able to automatically extract bits of information from unstructured or semi-structured sources, store them in a database, and offer a new search paradigm: instead of bringing back documents that contain selected keywords, we are applying database queries to deliver results.
“Whilst that approach is used by a number of companies in vertical search, we are applying our “secret sauce” to make our results as accurate as possible by pipelining a suite of algorithms delivering high precision, high recall results,” reads the website.
If the system works as well as they claim, they may have a major service to offer the entire web. Keep an eye on them.
