For many, contacting a customer service centre and waiting for them to reply is agonizing. While they place the call on hold, they sift through the relevant documents to answer the queries. The length of the call depends on how quickly they find information. It gets longer if they have to refer to bulky documents.
The city-based startup, Subtl.ai, sees an opportunity to solve this problem. It has created bots that will search the given documents to give the relevant answers in a quick time.
“We sensed an opportunity in the customer service segment. We worked on a tool and initially, we gave ourselves three years for the idea to percolate. But we have already got two big clients in about six months,” said Vishnu Ramesh, who co-founded Subtl along with IIIT-H faculty member Dr Manish Shrivastava.
“People need different varied information. However, in some cases the manuals are bulky. Our bots come handy in such cases,” he said.
It is now serving one defence related company where the bots are deployed to give answers on complex missile technologies. In another case, it has signed up with the big bank to give its bot. “The bank employees will be able to get the relevant information from the varied loan documents in real-time. This will result in faster decision making,” he said.
Vishnu said the bot can give the relevant answer even if the questions are reframed as it is based on ‘semantics understanding’. The bot learns with every query- it makes a relationship and allows the user to jump between documents.
The other use cases are auditing, due diligence, research, newsrooms. It has now created bots for banking, industrial machines and legal fields. “We can create a bot customised to new verticals,” he said.