It’s like the WordPress for Machine Leaning: Build ML models in five simple steps — all in a couple of minutes
Startup journalists write a lot about Artificial Intelligence (AI)/Machine Learning (ML) nowadays, even though this fraternity lacks sufficient knowledge in the domain. Lack of knowledge in tech at times is a blessing in disguise: A geek writer tends to write using a lot of jargon which can affect readability, whereas a writer like me, who has only basic knowledge in tech, always strives to present in simple language, in a layman’s term.
At times, ignorance is bliss.
As we know, AI is one of the best things to happen to the IT industry. Although it is always in an evolving stage, AI/ML has already sneaked into human intelligence, but its potential is yet to be tapped. According to experts, AI/ML will soon be part of every household, like the way computer did.
That said, like any other evolving technology, a key challenge is dogging AI/ML — talent shortage. Startups and companies across business verticals — from e-commerce to healthcare — across the globe are finding it hard to discover and hire AI/ML talent.
No more fretting about talent shortage
But this could be a thing of past, thanks to MateLabs. This Bangalore startup has developed a platform wherein one can build and train customised ML models within a short time in five simple steps, without writing a single line of code. So, no more fretting about talent shortage.
“In short, we are the WordPress for Machine Learning,” Mate Labs’s Founder and CEO Rahul Vishwakarma tells me. “Our platform Mateverse is a tool designed for companies who either don’t have a budget to hire AI talent or enough time to build the models from the ground up.”
Mateverse makes it easy for novices to use ML in their applications. Using a simple step-by-step interface, users can create and train a model which can easily be implemented into their programme using a few application programming interface (API) calls (watch the above video).
Here is how it works: A user comes to Mateverse, selects the appropriate datatype (images/text/comma-separated values), uploads file, and by just pressing the button ‘Start Training’, (s)he can get the model trained which (s)he can use it via graphical user interface (GUI) or API.
“Mateverse can be used by any business that generates a lot of data and wants to either optimise their current operation like loan defaulter prediction or financial budgeting using their own data, or wants to offer personalised service to their users like product recommendation. Using Mateverse, businesses can save tons of time which they otherwise spend on data scientists,” he adds.
It used to take a lot of hard work
The idea of Mate Labs came to Vishwakarma a couple of years ago when he was consulting companies in AI and ML. “We were consulting a couple of companies on how they can implement AI techniques to their problems when it struck us that there is a dearth of AI/ML talent in the whole ecosystem. It takes a lot of hard work for anyone to get a grip of the skillset in these technologies. In order to find a solution for the lack of talent and democratisation of ML, we were building an intelligent platform which could automate the whole process of building and training ML models,” he goes on.
Mate Labs Founder Rahul Vishwakarma
Launched in 2015 by Vishwakarma and Kailash Ahirwar (Co-founder and CTO), Mate Labs is targeting business professionals such as CXOs/vice presidents/business analysts who want to use the power of ML and data science to make smart decisions. A business analyst can use the platform to do a user persona.
The startup is primarily targeting e-commerce and chatbots using its technology. Currently, the platform has over 3,000 users.
Mate Labs has already raised US$50,000 in angel funding from Hong Kong-based AI accelerator Zeroth.ai and its Venture Partner, Sachin Unni. The startup is now looking to raise a seed investment to scale the platform.
How AI is going to change the world
AI/ML is still in the early stages, but it has already started changing the world around us, believes Vishwakarma. In future, AI and ML will enter almost every house like the way computers did. There’s a cycle for every revolutionary technology, which has its own maturity period. However, AI has been the fastest in that cycle compared to any technology in the past.
“There is a dire need for this technology to be democratised because even we, as an individual, are generating a lot of data which can be used to personalise our own digital surroundings in a way we want to. We use computers in our day-to-day life to augment our skills today. In a similar fashion, AI will one day help us reach superhuman capabilities, augmenting us to the next level,” he maintains.
“A service we launched in the similar line is an integration with IFTTT (an automation tool for creating your own if/then statements without any programming knowledge), and we are already seeing that happening. Now, a user can automate tasks which require an intelligent rationale. Mateverse acts like an intelligent filter. A popular usecase I can point out here is automating the hashtags when we post a picture online, or organising the photos on our phones,” he elaborates.
The company follows an AI-as-a-service model.
Huge demand from non-tech community
It is predicted that after a year or so, there will be a string demand/adoption of ML techniques from the non-tech community. “We have people coming to us and showing applications which an AI researcher/company has not thought of yet. There’s a huge rise in demand for self-service tools for business intelligence from business professionals,” he says. “This demand is exceedingly higher than what adoption of Internet saw. In the coming years, the adoption will come majorly from the non-tech community because they’re starting to realise its potential and maturing towards it.”
Mate Labs Co-founder and CTO Kailash Ahirwar
As for India, the country is fast catching up with the rest of the world in terms of adoption of AI/ML techniques. The fad which we saw two years back is now turning towards serious adoption, and companies are getting serious about implementing it rather than simply owning a web domain.
Mate Labs is currently working on a marketplace, similar to Google Playstore, where innovators will be able to sell their ML models to the ones who need it.
I asked: Do you think in future, AI will grow to a point where it can overpower human intelligence and emotions?
“It is a tough question. Personally, I feel it’ll be the same way how automobiles made distances look short. AI in future will be a supersonic jet for our brain. Whether they’ll overcome human intelligence and emotions or work hand-in-hand with humanity is difficult to predict,”Vishwakarma concludes.
The post With Mateverse, you can create Machine Learning models without writing a single line of code appeared first on e27.