Future Tech speaks to Janak Sarda, the CEO and Founder of Blue Logic, and Abhinav Kottalgi, the Director of Technology and Strategic Alliance at Blue Logic, about how companies can accelerate their digital transformation journey by using chatbots
Tell us about Blue Logic.
So we are a company that is about two years old. Right now, we are in a phase, wherein we are moving from a startup to an SME. That’s purely because of the kind of clients we have, the portfolio, the team size, and so on. We have now started focusing on Artificial Intelligence. The word AI is kind of a broad topic – so our specialisation is on two main areas of AI – we do conversation bots – chat and voice.
We also use NLP, integration, language detection. On the other side, we are heavily focused on computer vision. So on our computer vision side, which is the vision AI that we are developing, we have another set of product line which actually takes care of multiple sources as inputs. This could be in a video format, image format, or a document format. And we actually scan through that and get a lot of data out. So if you look at the image side of it, you upload a picture and we can recognise the different elements in there.
The data collected can then be used for content marketing, personalisation of information, and so on. Our next phase would be video recognition that includes face recognition, gender recognition, the emotion of the person, and so on. These parameters will help in retargeting the customer, enhancing the customer experience, and building a much better and efficient digital journey. So today, if you look at the healthcare sector, we can actually book a doctor’s appointment, end to end, with an SMS gateway, confirmations and so on.
Where does your development team sit?
Blue Logic has products for everything we do and I spoke to you earlier about. We have our own process of doing the complete automation. So, the chatbot product is something that we own – it is completely built in-house and our team sits in Dubai. This helps us address queries coming in from all the regional players. And this platform is multi-lingual. It supports languages such as English, Arabic, Bahasa, Mandarin, German and so on.
For the language processing, do you use your own proprietary platform or a third-party platform?
So we initially used to use Microsoft LUIS, which is their language processing platform. However, we have moved away from such third-party platforms, because on platforms such as these, you need to pay for each conversation. There is no fixed cost that the customer can really approve and hence the ROI gets hit every time.
Hence, we built our own platform and we now offer unlimited conversations to our clients through our platform. Developing our own platform gives us the leverage to customize it the way our clients would like. We can now integrate voice into our platform. So if someone does not want to type or read, can speak to the bot and the bot will speak back to you with a response.
Do you have plans to launch bots for other industry verticals as well?
As a company, we are vertical agnostic. We don’t mind working with clients from verticals other than hospitality or healthcare. When we started off, our very first client was in the healthcare industry. So if a new healthcare client comes to us today, we have lots of pre-trained data sets that can be shared. Thus, it becomes easier for us to develop bots for the healthcare industry.
We are also doing lots of work in the insurance sector – so there, it becomes easier for us to provide a bot that is pre-trained for that sector. We are also working on our first education bot, for which we don’t have any pre-trained data sets. So we are now trying to build and train data sets for the bot.
What sort of response have you received from your customers and how do you handle all the data that is collected and processed?
The response has been very positive from our customers. Most of our products are cloud-base, however most data is stored on-premise at the clients’ facilities. We don’t store anything – most of our chat bots are on AWS, except for two bots which are on-premise. Now when it gets on the cloud, a question arises as to what sort of data is it and where is it stored. So, we store two things. We store the shat transcripts, because our learnings come from chat transcripts.
Now, when it comes to sensitive data, it is never through the bots. It is always through forms. So we ask people to enter their phone number, or the date of birth inside a form. The data is then encrypted, masked and sent back to the clients’ systems through an API.
So what are the plans for 2020?
We are working with lots of brands in the region. In fact, we will launching our first voice bot for the region this year, specifically focused towards the hospitality sector. And the bot will support Alexa. So you essentially will be able to book a table in any of their restaurants, just by speaking to the bot, from anywhere in the region.
We have close to 15 bots that are live already and some more are in the development phase. We are also working with a large e-commerce player and they will soon be going live with the first e-commerce bot in the region. We are already working with companies such as Aster, Medcare Hospital, Sony Middle East, National Bonds, and so on, and we plan to launch more bots in the coming months.