Tiger Analytics, a data sciences and advanced analytics company, has undergone a rebranding exercise. Storyboard18 caught up with Mahesh Kumar, founder and chief executive officer, and Pradeep Gulipalli, co-founder of Tiger Analytics, who spoke about the rebranding initiative, the combination of AI and analytics, and the benefits of generative AI.
Rebranded logo
Could you touch upon the rebranding initiative of Tiger Analytics?
Mahesh Kumar: The last time we may have done something similar on a much smaller scale was when we were a team of 100 people, which was almost eight to 10 years ago. Today, our team has grown to 4,000 people. Now, we have multiple large offices in India and seven to eight countries.
We provide services in areas including data science, data engineering, ML (machine learning) engineering, MLOps (machine learning operations), quality engineering and application engineering. We are doing quite a lot of work in generative AI and large language models.
Today, we have 100+ clients. These are large companies spread globally, and the nature of the work spans across retail, consumer product goods, pharmaceutical, insurance, banking, transportation, logistics and manufacturing.
We have been constantly evolving our messaging when we interact with the external world, our clients, agencies as well as internally. But we felt that this is the time we need to take a holistic look at what we are trying to do here, create a consistent simple story to convey to everyone in this complex world.
What are the goals you aim to accomplish through this exercise?
Pradeep Gulipalli: What does building the world's best AI and analytics firm mean? What do we do? This is what the whole branding exercise was about.
The whole business ecosystem has gotten pretty complex. You don't know a particular action any business takes and what the outcome will be. Customer behaviours are changing with all the technology and the revolution that is happening. The competitive landscape is also quite dynamic. ChatGPT didn't exist a year ago and then see how dramatically things have changed. So it's a pretty complex situation that our clients deal with.
And how do they make decisions? Sometimes they have a lot of data. Sometimes they have no data. Signals are hidden. It's ambiguous. It's not easy. So many times, they end up just sitting on it. This is where we come in and we say, We will make sense of all of this for you.'
Generative AI is a big buzzword. How are you making use of this tool in your daily practices to solve the toughest of problems?
Gulipalli: If you take a look at our work, we are working on some business problems. It could be trying to predict something, some forecasting or trying to optimise something. Now, generative AI is doing two things for us.
One is that the process of developing these models is becoming easier now. It's been heard that a generative AI can now replace software engineers or it's able to write code. Parts of it are definitely happening. Not the replacement, but certainly it's substantially improved productivity.
But that's one part. The second part is solving actual business problems. Things which were unsolvable earlier, now some solutions are coming up. There is a live project we are working on with a very large company. I consider it to be among the top three BPO companies in the world. They run call centres across multiple industries globally.
We are working with the group that works with airlines. With airlines, it's easiest to buy a ticket. But then once you ring the call centre, it can be a frustrating experience.
So the solution we are building for them, we're doing it in two phases. Phase one, when a customer calls, they are going to listen to that conversation live with the help of AI.
And then, there are generative AI models which understand it. And they are giving a solution to the query being raised by the customer. Now, phase one was, Will there still be a human who will answer?
We are giving that solution to the call centre agent, saying, Here is the answer to that question.
Now, in the next phase, we are saying, 'Can we experiment where the generative AI or the bot directly answers the question, without a human in the loop?'
And there will be some pockets which are complex enough that only humans can do. And people can spend more time there. So that's phase two. So this is an example. And you can draw parallels to how other people would be using the massive amount of information available, but we don't know what is relevant. That's where AI comes in handy.
Kumar: About 1.4 billion people in India can register their complaints with the ministry of consumer affairs. Any product or service you buy as a consumer, you become the customer for the country. Hence, you can register your complaints.
And earlier there used to be fewer complaints just because the channels were cumbersome. Now they have opened up mobile channels, email, phone, etc. So, the current government is encouraging more registration of these complaints, suggestions and feedback.
The problem for them is now the volume is so high. There are only 25-30 people who are manually processing those complaints right now. So, they are launching an RFP right now for the use of gen AI for processing these complaints. And the purpose is categorising them.
First, many complaints are duplicates from the multiple channels people are getting. So how to deduplicate them? Second, some could be bogus complaints. How can they classify them into relevant, non-relevant, depending on the intensity of the complaints?
There's a question: How important is the complaint? Everyone says it's an important complaint, but they do internal categorisation. Then also, which category kind of complaint it is.
So all of that can be done by gen AI, which was earlier done manually. Eventually, what happens is if the complaints are more serious, one can approach the court. What they are trying to do is, can AI provide an intermediate solution, almost like negotiation between the service provider and the consumer? So there are a lot of interesting applications coming up with gen AI, which is going to touch the day-to-day life of every individual.
What trends are you expecting for 2023 in the space you cover? And could you explain the disruption that you will witness in space?
Kumar: Tiger Analytics was started in 2011 and since then, we have witnessed a lot of changes in the work of traditional data science and data engineering. In the post-Covid era, the demand for and adoption of these technologies has increased significantly with the entry of cloud.
The demand for gen AI has increased, where it led to the increase of our services. So gen AI is in that category where it has a promise of high value.
Earlier, if one went to large Fortune 500 companies, there would be 100 people talking about artificial intelligence (AI), data science and data engineering.
But gen AI has everyone talking about it. So, it has expanded the total addressable market of the services we provide. That's a big change we're seeing. And it will translate into much wider adoption of what we do.
Through AI and analytics, what are the challenges and opportunities Tiger Analytics has helped companies overcome and embrace?
Gulipalli: AI and analytics mainly help with decision making. Coming up with decisions based on data is a much more scientific way of making a decision. Sometimes the decision can be manual or it can be automated. AI can make the decision for you. But at the end of the day, we are making a decision.
Now think of looking across the organisation. You have a product which you are trying to sell. So it starts with sales. How can I better sell? Who is the right audience? How do I sell? Who do I market it to? How much do I spend on marketing? Today, there are a number of ways a product can be marketed.
There are hundreds of television channels. Digital has exploded. Where do I market? AI can give you answers for all of these. For example, you have acquired a customer. How do you provide a good customer experience? How do you retain the customer? How do you address their pain points? Identifying issues manually is a huge problem.
AI can help identify these hotspots. Then, you approach the operations of a company. Are you having optimal usage of resources? Is there wastage? Through AI and analytics, there can be better optimal usage of resources.
When you try to do planning as a business, you want to know how much I manufacture. For that, you would want to know what the demand is going to look like. How do you predict demand? There could be so many things that impact demand. The decisions spanning from sales, marketing, customer, operations, financial, risk, manufacturing, supply chain, all of these are things that can be solved using analytics and AI.
What makes AI and analytics a very deadly combination? Is there any other tool you would combine with analytics or AI?
Gulipalli: AI is like an engine. Think of it as if you're building a Formula One car. AI is the engine. But then just that engine will not get you to your destination. You need to build the overall chassis, have the steering, the seats, the body, etc.
There are different technologies which will come in to complete the picture. For example, AI is great but let's say for you to understand what AI is saying and for you to act on it, maybe you need a certain technology which is an interpreter between you and AI. So there are a variety of adjacent technologies which come in and help. And there'll be more that will keep coming in.
What AI was 10 years ago is not the same today. A lot of advancements have taken place. A lot of embellishments will come and enablement will come in terms of other technologies. So we have for a long heard about software development. So, both are very much tied together.
This is because, at the end of the day, what you'll use is a software product but at the heart of the software product is AI. For you to understand something much better, maybe you might need certain visualisations so that you can grasp the complexity of a situation better. Now, this is where AI and the visualisation technology will work together.
Currently, how many clients do you have?
Kumar: Right now, we are working with 130-140 companies. Seventy percent of our clients are from Fortune 1,000 companies.
In the consumer product goods category, among the top 10 companies, we work with six of them. Majority of business is in the US right now. We do have some business in Australia, Asia, Singapore, the Philippines, the UK and Malaysia. We recently started working with some government entities.
In India, we work with Tata Steel and with Star Network at Disney India. We are just starting our first project with the government of Bihar, which revolves around AI. In the US, we have clients like PepsiCo, The Hartford (The Hartford Financial Services Group) and Nestle. We are about to start work with Citibank.
Continued here:
Tiger Analytics: Generative AI has improved productivity - Storyboard18
Read More..