The next frontier of Automation: Generative AI in action

Prankur: Hello and welcome to an all new episode of the Zinnov Podcast, where we dive deep into the latest trends and innovations across the world of Intelligent Automation, getting unique perspectives from industry experts, thought leaders, and technology practitioners.

I’m your host, Prankur Sharma, Principal – Automation and AI at Zinnov. The Intelligent Automation space is currently ongoing a transformative phase characterized by rapid innovation and widespread adoption. Organizations across industries are revolutionizing the use of potential automation technologies and outcomes around experiences being enhanced and operational efficiency at an exceptional pace. As the technology landscape evolves, especially with the advent of Generative AI, the focus is shifting towards human-AI collaborative approach with seamless integration and robust privacy and security. To talk about this, today we have with us Vijay Thomas, the CEO of Tangentia. Vijay founded Tangentia in 2003, an automation solutions company headquartered in Toronto, Canada. Vijay is an autonomous enterprise evangelist who believes that with the right skills and technologies, nothing is impossible. Welcome, Vijay. We are thrilled to have you.

Vijay Thomas: Thank you very much, Prankur. And I’m really thrilled to be here as well.

Prankur: Great. Let’s get started. So we understand Tangentia is one of the leading automation service providers in the boutique automation category where we are tracking almost 200 companies in that sub-hundred million revenue range and these companies solely focus on automation.

Can you share your perspective on what gives an edge to boutique firms versus some of your larger and mid-tier counterparts and what really helps you set apart in this competitive market space?

Vijay Thomas: Absolutely. There’s a handful of large global consulting companies. And then there’s the GSIs which are Global System Integrators. That’s probably another handful or maybe a little more than a handful of GSIs. And then there’s a whole plethora of boutique firms.

And a lot of these boutique firms are either geographically focused in a certain market, or in a certain industry and they are usually run by people that come out of that industry, or have skills in that geographic area. So they built a practice around that and they do pretty well.

So we at Tangentia look at ourselves slightly differently. We think we have something very unique and we call ourselves global boutique. Now global boutique sounds like an oxymoron. How can it be global and boutique? But essentially that’s what we’ve been able to pull off. We have a boutique delivery focus with a global delivery model.

So we’re local in a lot of the markets. We have operations in Canada, operations in the USA, we have operations in India. We also have a small operation now in Mexico, but our delivery is global. So, we actually have the same model for delivery that the large global consulting companies have, the GSIs have, and what we’ve been able to do is bring the global delivery model to smaller size projects. If you go to a large global consulting company or a GSI, they wouldn’t touch a project in a global delivery model less than maybe a million dollars, in some cases even a couple of million dollars.

But if you are a line of business or a mid-sized company and you want to do a project for, let’s say half a million dollars, or even two hundred thousand dollars, or a hundred thousand dollars, they won’t touch that. But a company like us we can do that with a global delivery model with the governance that is required.

So we’ve built our governance models for smaller projects. If you go to a Deloitte or a Capgemini or an Accenture, for them to put together a PMO, the size of the project itself has to be quite significant. So I think that’s our true differentiator, being able to do this. So, there’s boutique in this global and there’s global boutique. And we like to think we’re kind of leaders in that spot.

Prankur: Yeah. Wonderful to know about this new category, which I think global boutique, it certainly has a ring to it. And shifting gears a bit and coming to the flavor of the season Generative AI. I think that’s the buzzword in the automation space today. What is your perspective on how automation and Generative AI coming together, especially in the context of automation, which functions, industries do you think are witnessing the maximum impact from generative AI?

Vijay Thomas: So Generative AI is here to stay. We can fight it or we can embrace it. I’m of the view that the world has already embraced it. If you as part of your business or any enterprise has not embraced it, it’s just a matter of time that you will have to embrace it. So it is inevitable. I think it’s been happening for quite some time, but thanks to ChatGPT, it’s really got everyone’s attention.

In fact, I would go as far as to say that maybe someday in history, we might actually have a timeline, which is before ChatGPT and after ChatGPT. I mean, it’s probably a little out there, but it is possible, ChatGPT per se may not be, but ChatGPT is representative of a whole lot of different technologies. It is kind of the flag bearer for all of that. So it kind of put all of it together. And with automation, think of it as what do humans do when we work? There’s some repetitive mundane tasks that we do, and we repeat and do those things.

Guess what? All of those things automation kind of covered. Now there were things where you needed to be a little more creative. You had to come up with something. You had to write a sentence, I’ll give you an example. We’ve been working in the  e-commerce space for almost 20 years.

So if you went to a company and they wanted to start an e-commerce site. And you got these images, you have to tag these images.  I still remember the time, 10 years back, we would get a bunch of interns, and they would manually tag images. And then what happened? We trained an AI model to write those tags. So AI model knew enough to write those tags. Now, with Generative AI, in that space, you still needed a copywriter who’d write a little bit of say, imagine yourself riding a bike, wearing these denim trousers. So that part was still human, even after the tagging was automated. Now, with Generative AI, guess what? That story as well can be Generative AI. So, Generative AI is going to take on more of human tasks. You can have either through some large language models, you know, things out there, you can incorporate some of these things. So I think it is going to happen. Do we really need humans to write a script for furniture, for things like that. Potentially not. I mean, you know, the jury is out there and we can we can talk about it. But, do we really need people to do that? I mean, so I think that’s where I think the evolution is.

And exactly on e-commerce. I think this is an example I like giving, because e-commerce, you can know where automation came in. And actually it’s some bit of machine recognition and visual recognition and all of that stuff came in and then the Generative AI. So all of this comes together, I think this is a good example of how everything that you see on websites and stuff, think of it as a site like Amazon which has millions of SKUs.

And, and today some of those SKUs, even the description is not very good. So I think there is a school of thought that says it might get worse, but you know, today it’s quite bad because there’s no people to write that. So if you can do this, you might actually get into a better spot. So I’m on the side of Generative AI. If there’s doubt, I think guardrails and other things are another discussion, but I think there’s a big gap that can be filled in terms of increasing quality and doing things better.

Prankur: Definitely, Vijay, I think a lot of food for thought for us and for technology practitioners who are looking to leverage the technology. And just going a bit deeper into this, I think you have been using conventional AI for a long time. You touched upon the whole e-commerce example and the technology stacks, from the hyperscalers, including IBM, Google, Amazon is something that you have been leveraging for a long time.

In the Generative AI space specifically, we see while all of these hyperscalers are making tons of investment, they also specialize Generative AI as startups that have come up. How do you see your partner ecosystem evolving? Is there going to be a transformative shift in terms of where you get access to some of these capabilities?

Vijay Thomas: Yeah, very good question. Some of this keeps me awake at night because as a company we’re not a deep technology company. So we interface with the customer. We make sure that we use the deep technology to create value for our end customers. So, I give the example of us being like a kayak. Have you heard of kayak? And most of you have heard of kayak. It’s the meta search engine. So Kayak does not really do all the booking or anything. It’s a thin layer that runs on top of Amadeus and all the other booking engines, whatever.

Kayak probably has a higher market value than all those booking engines and everything put together. So anybody who controls the customer and actually provides the value to the customer is the king. We at Tangentia would like to be in that layer where we are leveraging the hyperscalers. We are making sure that we are using those technology stacks, all those large investments that we have done to make sure that it is available to end customers in a manner in which they can consume quicker, they can get to value quicker.

That’s where we come in. I think with the hyperscaler model you won’t go wrong. You can definitely look at it and then the delta between a hyperscaler and a niche player, if it’s a huge difference, then obviously you know, people should consider it.

But again, for a 5% increase in effectiveness, does it matter because the hyperscaler ecosystem comes with a whole lot of other things as well. So it’s not just that point solution. So I think there are trade-offs and we will see trade-offs becoming more and more relevant in the automation space specifically, we are seeing. Microsoft making big investments, and Microsoft owning the desktop today globally. It has become an easy decision for companies to say, you know, we’re going to try the Microsoft Power Platform, we’re going to try Power apps, we’re going to try AI builder. We’re going to try some of these things that come out of the box with Microsoft, you know, do we need best of breed and is the best of breed so much better? So I think the hyperscalers do have an edge here. People are able to extract a lot more value for what you’re doing.

Prankur: Going  deeper into the debate, we understand, I think, the shift is happening between processes that were just getting automated to now, I think, almost autonomous workflows and I think it brings us to that fundamental question on how much of this whole decision making that process will be owned by machines versus, what will humans really do in that whole enterprise functioning? So, what is your opinion? How do we think of this autonomous future?

Vijay Thomas: Again, very good question. And I think lot of this is going to be philosophical. I mean, it is going to be where, you know, today in the car industry most portions of the car are made autonomous, you know, very automated robots, quality control, everything is automated.

In India, Maruti Suzuki probably makes a car every one minute, or two every minute, they are making millions of cars, right? However, you know if you go and buy Mercedes-AMG, the AMG motor is hand built today. Why is it hand built? It’s hand built because there are some people that see a value in hand built things. So where I’m going with this is today the options for companies to what to automate and what not to automate, where to have humans, not to have humans is going to be entirely theirs. There is the option for them to automate everything if need be.

Technology is not there as yet to automate everything, but at one day technology will be there to automate everything. Today Mercedes-AMG can automate the making of their engine as well. But they have not done it, because they are positioning that as value because it’s human made. It’s hand built and somebody signs it the actual guy, one guy builds it and signs it. They charge you a premium for that. There will be companies that will say we use human labor or we use humans and they will charge you a premium and people will go to that company. The other guys say we’re fully automated and we’ll give you best prices because we’re fully automated. So the dust is yet to settle on how this is all going to evolve. But i’ll give you the example this Maruti Suzuki and there’s AMG, a hand built motor.

So the answer will be somewhere in between. So today technology cannot do everything. So today anybody says we are fully automated is also bluffing It is not. The technology is not caught up till there. We we’re still evolving. But a few years from now, it will almost be the decision of the companies to figure this out. I think if we live in interesting times, there will be people today buying handmade soaps in India. Why in today’s world are people making handmade soap?

Why? I mean, hey they’re charging a premium for it. So this is going to evolve, but you don’t need to have humans, but you will still use humans for multiple reasons. And some companies will want to do business with companies using humans. And, it is how the world is. I mean, there will be options for everybody.

Prankur: Yeah, I think that’s a good way to put it, there’s certainly a premium attached to handmade stuff and I think probably that’s how that will also be a consideration for how companies think about operating their businesses going forward. You spoke about a lot of examples and use cases of how automation and AI are being used within the enterprise for driving efficiencies, for exceptional experiences, for empathy within various functions of companies. Is there any, uh, specific example of intelligent automation that is really close to your heart in terms of either the problem being solved or the outcomes that were delivered?

Vijay Thomas: Yeah. Something that I thought was in the non-profit business, you know, sometimes, you gave away money, you do something. I was at an event in India. It was the Pravasi Bhartiya Devas. It was in Indore and you could plant a tree. Okay. You could plant a tree. You didn’t actually have to plant a tree, but you gave a donation or whatever. And they gave you a QR code. And then, and this QR code kind of connected to a network of computers or whatever.

So at any time during the year, you could actually go check on your tree, do a bunch of things. So I was thinking about it. So this is fantastic. I mean, you could look at growth of these trees. You can see how much carbon monoxide you’re taking out the world, how much oxygen was being put in. You help build a school in some area, you know, you can actually measure through AI how many people are attending school, are the same people coming in. So this is the good AI kind of portion, I think is where you can actually ensure that whatever is being done is also monitored, course correction can be done quickly.

Sometimes even autonomously can be done. If your tree died, you can immediately call for a new tree, right? I mean, and it could be automated. You don’t need for somebody to go and do a check of all these things. So I think this is a lot. In the non-profit sector, in the ESG sector, I think there is, a lot more case studies of how this can be used.

And I think, we can make the world a better place. Obviously the enterprise part is we’re all taking it for granted. But I think in all these other parts as well, there is the use of technology… is definitely something that I think can provide immense value. And it almost will make that whole investment more efficient. It can make the non-profit sector more efficient. People are going to be, you know, can see their money actually doing work. And I think, it could be a game changer.

19:36

Prankur: Yeah, I think that will certainly give a big boost to philanthropy.

If I think we can successfully deliver on some of these promises. I think in the same way that you spoke about it. Thanks for sharing your perspectives. I think we learned a lot about what is the possibilities around Generative AI. How is the technology ecosystem evolving? Promises of generative AI will certainly lead a lot of companies to invest or at least give a fair chance to the technology and you’ve left us with these interesting perspectives to think about, and I’m sure that our listeners found this as engaging as I did. Thanks  once again for taking time to do this with us

Vijay Thomas: Thank you very much, Prankur. Really appreciate it. Thank you for having me on your show.