Clarifai is traditionally known as a web-based visual search tool that developers can integrate into their services. But as more and more businesses begin to access the service, with their own specific demands – like getting them in-house – the New York start-up has to grow a bit.
Today, the company is starting to take some early steps to get there. After raising $ 30 million in October 2016, the company announced today that it will begin rolling out an on-site solution for large companies, where they could use all Clarifai's services outside of their own servers. . Large companies often have their own proprietary types of data and clients to work with, which have a myriad of different demands, and sometimes have to make sure that their customer data does not monitor them closely – or that it has to do it as quickly as possible. t allow the delay to connect to the cloud.
"Think of a casino with 10,000 cameras all going to a central location," said CEO Matt Zeiler. "You have like a streaming service that [faster than] runs through the cloud that must be close to the edge.The camera is there, it's not enough, so you have to broadcast it on a hub You need to reduce latency, increase latency, and protect privacy.These are all the same API objects and structures that you would use. "
The company also said it has made two hires to senior executive positions: a vice president of engineering and a vice president of products. Ulas Bardak joined Clarifai as head of engineering after working at Whisper, while Rajesh Talpade previously served as product manager for Google . These two hires are part of this growth process as he begins to try to entice more companies – and the kind of talent he needs to build these tools.
Part of Clarifai's presentation is that it can be a neutral celebration with regard to image recognition. While companies like Google and Pinterest can have their own data sets and build custom visual search algorithms for their products, Clarifai is working to build these tools that other companies can leverage to have their own visual search tools. This is part of a process of transferring some of these processes to a third party and focus on the key elements of their products, a feature of Web development that has led to an explosion in the ecosystem starting.
"When you use something like us, our research team continually improves the algorithms, and continues to collect data, which will give you a higher level of accuracy and improvement over time." declared Zeiler. "This is a huge benefit and it avoids the speed at which you train, we want it to continually improve.That takes a lot of resources and data, but it's the big advantage of signing up for a service like us.We always compare them in terms of quantitative elements like accuracy and qualitative elements like the team that reviews the models before publishing them for s & # 39; Make sure they do not make stupid mistakes. "
To top it off, Clarifai also makes a small image of the site. It's a small detail, but all of this adds up to ways to help companies express exactly what they're doing and to create the tools needed to become a must-have tool for developers in research. visual.
Clarifai, of course, runs the constant risk of some of these bigger players and the chance that they are just going to open up their visual search algorithms for developers. There are obvious candidates, but at the same time, if they are trying to build a walled garden, they probably want to keep those data and tools for them. If this remains the case, Clarifai could buy enough time to collect enough data, have all the elements of a given image or video tagged, and offer companies a sufficiently compelling argument for that. they adopt it.