Ep 47: Innovation at the Edge: How AI Transforms the Enterprise with Bain Global CIO & CTO Ramesh Razdan

‍Evan: Hi there, and welcome to Enterprise Software Innovators, a show where top tech executives share how they innovate at scale. In each episode, enterprise CIOs share how they've applied exciting new technologies, and what they've learned along the way. I'm Evan Reiser, the CEO and founder of Abnormal Security.

‍Saam: I'm Saam Motamedi, a general partner at Greylock Partners.

‍Evan: Today on the show, we’re bringing you a conversation with Ramesh Razdan, Global Chief Technology & Information Officer at Bain & Company.

Bain & Company is one of the “Big Three” management consulting firms, with 65 offices around the globe, 19,000 employees, and over $8 billion in annual revenue. Bain helps Fortune 500 companies optimize performance, enhance operational efficiency, and implement innovative technology at scale. In this conversation, Ramesh shares his thoughts on how generative AI is transforming Bain, the evolution of the CIO role in the AI era, and how Bain’s clients are harnessing the latest AI capabilities.

Well, Ramesh, thank you so much for joining us today. I know Saam and I were really looking forward to this episode. Maybe to kick us off, do you mind sharing a little bit about kind of your, your background, your career and how you ended up in your role today?

Ramesh: I've been in high tech company for 18, uh, 17, 18 years. Then in the management consulting for the last six, seven years. Prior to that, I was in a training and education company, so I have had, uh, from high tech to manufacturing to professional services, different types of roles. I think what defines me is always, kind of meeting the cutting edge of the technology, helping drive that and more importantly, driving the full potential of technology can do for business.

So I think it's been a fascinating journey. I have gratitude to what I have been able to accomplish. But more gratitude to the leaders and, uh, people who actually helped me reach what I am today. 

Evan: And you, you've worked at some really notable companies, right? Dell, BCG, um, do you mind sharing a little about the role you're in today and the type of things you do to kind of drive technology innovation at Bain?

Ramesh: My role today at Bain is I'm the CTO and CIO for Bain & Company, which basically means everything related to technology, and it has three aspects of that. One is obviously, how do we run, uh, technology at Bain, uh, internally, uh, supporting Bain & Company operations, what types of products and services we offer to our clients, I'm responsible for that as well, and more closely working with our technology practice, which we call Vector, as well, that's mostly, I would say, inspirational, emotional, uh, thing to how do we help them as well.

So I, those are the three key roles that I play for Bain & Company. 

Saam: Ramesh, one of the reasons we've been so excited to have you on is you've been a part of the technology ecosystem for several decades from a number of interesting vantage points, and you've seen a number of innovation waves. 

I think it's hard for us to be talking in 2024 and not be talking about the technology wave of our era, which is around AI, right? And so I'd love to learn more about your view on this wave, where we are, how's it similar and different from other technology transformations. And then we could double click on perhaps Bain in specific. 

Ramesh: Yeah, I think glad to share my point of view. I think it's not anything that you and a lot of cases AI has been there for a long time. What is net new is generative AI, which is the ability for us to summarize and synthesize some of the information.

And I think that's where people get lost. To me, the value disproportionate value gets created when you merge generative AI with what we call is predictive and prescriptive AI. I've been lucky enough to run a data organization and analytics organization for the last 15-20 years and done a lot of things that actually did not need generative AI part of that.

And I think the question is, why now? And, like, what is the value it can create for humanity at large? And there are fundamentally four key reasons that why now. One is the abundance of information in the internet, there is way too much information. We have way too much information, but no insights and, um, all over the place.

And I think the second part of it is, I think the computer is getting more and more cloud computers available. The availability of GPUs is more and more relevant to the industry. So I think we can do it more cost effectively. Although the curve has not completely been where it is more affordable to the world than ever before.

I think third important is the architecture. Uh architecture what was done, which is started with the transformer architecture. To me, that's like the thing that actually has helped a lot because now we can do it as a scale and do that, do that broadly. So to me, it is a culmination of Internet has been there and now Internet has created a lot of information. We have a technology that can do that. We have an architecture that can do that. And we have a scale that can do that. So it's all a culmination of that. Why now? And because if you are trying sometimes technology is ready, but you can't be consuming it, I think we're still going through that. So what is different about this is the ability for this to kind of, be like human.

As I said, we have way too much every enterprise is overwhelmed with way too much information and synthesizing that and simplifying that is, is a big challenge. And then kind of deliver insights from that. I think there are a number of use cases and we can get to that. I think that there is another, you can look at it from a two, two aspects.

One is how do you create more value? How do you think about an enterprise? To me, there are only three ways to create a value. One is, what products and services can you offer? How do you create new service lines? 

Then the second part of that is, what are you driving from efficiency standpoint? What do you do for bottom line? Whether that is productivity or whether that is efficiency, simplifying certain things.

I think the third one, how do we in this world of complicated risk and security, how do you actually deal with security landscape? It is such a nightmare to deal with that. Being an IT professional today. We have too many facets that are open and it is impossible to know which facet to close and which facet to deal with. So, I think this is an opportunity to address all three of them, which is pretty unique.

And most technologies either play in one or two. To me, this allows us to do in all three aspects.

Evan: Are there, kind of, a couple examples of ways you guys have deployed technology that have really kind of helped the business, right, in the past. There are things that maybe even your, your clients might be surprised to hear about, you know, some of the internal innovations and, you know, unique ways you guys do things.

Ramesh: Yeah, I think we are uniquely positioned to help deliver extraordinary value to our clients. I was using the word extraordinary. We say that we deliver extraordinary results to our clients, and that's our goal. And I think when you think about ChatGPT unfolding to the marketplace in like November of last year.

We'd be surprised know we are actually working with them six months in advance. We have three use cases ready to go. And even the world did not know that, that the gen AI kind of us. And this helps us because we are connected with the private market quite a bit. We have a private equity, uh, one of our most, uh, strong positions is, uh, what we do with the private equity. So we're working with already with openAI before OpenAI was kind of explored the markets. We have three use cases. 

So I would say we have been lucky enough to be part of the journey. We have actually delivered a lot of use cases. I think some of the use cases maybe I can show we have 15 plus products today live in production and 100 plus innovations that we talk about.

So I'll talk about two lenses. What I see from a client standpoint and what, what I see from internal standpoint. Internally, I would say the number one use cases for this, at the end of the day, we're a knowledge company. We have built a chat bar that we've built like almost year plus now has been as a NPS and everything we deliver on look at being invented NPS as well is like we major in NPS. 

NPS means net promoter score that anybody uses a solution is actually a promoter. We have an NPS of 80. The people love the solution. We are able to do that. And the key thing is not only to bring the information, but give it to the reference and give me the full lineage of where the information came from. And that's mostly internal information and the public information obviously is out there. To me, that's like one of the big use cases. 

We're building one use case for our performance. You look at some of the things that people don't like to do is reviews. Reviews and performance reviews. How can you simplify, streamline, and bring the performance reviews part of that? We are actually building a use case for that, which we did the version 1. We're not thinking about version 2 of that. 

Then we have use cases, like some efficiency use cases would be a contact center use. How do we simplify the knowledge and empower the users?

We, at the end of the day, hired smart people. We want them to be productive. We want everybody to be reducing the friction. My goal is always deliver a world class experience, reduce the friction in the system because people will latch on to something that actually reduces friction, particularly the friction they don't like. So we have something like a contact center chat bot that can answer some of your common things that you can do there. 

And then there are a number of other use cases. We have a recruiting chat bot. We're using tools like Zoom, Zoom AI and whatnot. There's a lot of short tools. We're using OpenAI's Enterprise ChatGPT. We have roughly 3000 GPTs today, that actually all deployed. So we have done a lot. 

Evan: As you talked about, you know, thousands of GPTs being deployed, can you share a couple examples and maybe ones that might be unique or surprising, right? Where there's kind of use cases that you guys have seen some success with that, um, you know, maybe other people wouldn't, wouldn't expect, um, wouldn't expect kind of these custom GPTs to be as effective.

Ramesh: Yeah, I think there are a number of cool use cases that we have done. For me, personally, I deployed it like last year, I have an email chat, uh, email GPT for myself because I want to reduce the friction for me writing. I want to give a context. This is what I'm trying to be aspiring for. And I think more and more people like for executives and other people, how can you synthesize and simplify the information, is a key thing. And but I want the GPT writing like me, not so I have to give it the context, and I have to educate that so can that is definitely something that I use myself on a day to day basis.

We started the training from our board. We trained them to say this is what is coming down and you should all be aware about that. Some of them use a day to day basis. Some of them Uh may not be using a day to day basis. And then you can think about A number of use cases. 

I think you can see even the people that the power of the technology is, it gives you the power for, it's a low core platform, a lot of cases. You can look at people thinking about how to build, how to show the thing that is being done in a Python. And so you can make changes. So a lot of our smart people have actually, kind of, updated a model, something that would have needed a specific python and code. They actually done it coded in the GPT, so we have a GPT that does updates certain models on on the regular basis. 

We have a recruitment bot that we have built recently. We have a service synthesizer that actually you can synthesize so many surveys and actually build a common. I don't think we have explored every opportunity that is out there. I'm sure there will be many, many more coming down the pipe. 

Evan: Earlier you said, um, one of the, like, I guess, philosophies you're trying to enable is innovation at the edge with kind of centralized orchestration. So again, I'm just still very impressed with the kind of deployment of kind of GPTs at that scale. 

Do you want to share a little bit about how you're able to do that? Right? What guardrails do you put in place or what enabling do you have to do to enable, you know, all of the users that are maybe the experts are close to the use cases to go kind of take advantage of that, of these new technologies, with the appropriate kind of structure and governance around it to make sure that people aren't doing duplicative work or kind of crazy, inappropriate things like, how did you kind of like organize that? Because I think there's probably a lot of people going to be listening to what you said a minute ago, being like, wow, I wonder how they got that in place, so quickly in the grand scheme of things. 

Ramesh: To me, the foundational thing is understanding the use cases, giving few examples, Showing the art of possible is critical. To me, that's a foundational enablement step. 

I think the other part is you got to inspire and motivate the, the edge, I call it. That, okay, this is the use case. You have a power to in influence vein, you have to power to influence what you can solve as, as an individual and as a collective team. So to, to me, that is another part. 

We are training, educating. Asking people to do that. We're not done, to be quite honest. We are actually still going through the process because we want to train people who want to educate them.

I think the third one is codifying the controls inside the systems. Before we rolled out, we put controls in place that you can't retain any of the data. The data will be deleted automatically. So making sure that security is built into the process. So you, when people come in, because you can't ask people for security, they want the things to be coded in the process. What can you do whether you lo or launch Zoom, whether you launch, uh, ChatGPT, what configurations do you put so that you minimize the risk to associate you to that? I'm not sure you can completely eliminate a hundred percent of it, but what controls? We put a disproportionate amount of time to do that. 

I think that also comes back. We also know we are in the early innings of this AI, whether we are in early or early to midsize, how do you build an architecture that enables you to think about how do you build it so that you can replace some components, we want to do that. Like, if there is tomorrow, a better model to do that, can I leverage that? And we have seen that. 

We have seen also that something that we innovated a year ago today, out of the box, is available in some of the tools. How do you sunset some of that process? We are actually about to sunset a tool that we built, and we were able to do that. We needed a code to do that. Today, out of the box comes from some of the tools. So I think that innovation is only as good as if you sunset some, otherwise, there'll be too many things to do that.

We are still, still learning, to be honest, but I'm excited and energized what actually we can do for, uh, for Bain because in terms of harnessing innovation and solving some complex problems.

Saam: So I think one of the things you can help illustrate for our listeners, given the vantage point you have is, without naming a specific client, just what are some of the most interesting ways you've seen clients evolve and transform their businesses, leveraging AI? 

Ramesh: What's fundamentally has shifted in the industry to me is technology is the disrupter now to the industry. Techno rather than technology being an enabler. Oh, I need to run a RP. I need to run a CRM. Technology is disrupting the business. Which translates to me that leaders have to understand what I say is one of the key. My personal reflections is, leaders need to have one of the key skills is blend strategy with execution, because strategy without execution doesn't work. 

So I think a lot of people are stuck. Oh, how do I keep my SAP up and running? And how do I do this stuff? That is important. That was important yesterday. But today it is, how do you help transform the business? And how do you actually connect to that? And sometimes we're not able to do that. So I feel like that's, that's a Uber thing that I think in this new world, in new era of technology disruption is a key and a key skill, uh, uh, technology leader needs to have and we can go to a lot of stuff. 

So I think in terms of the use cases I have seen, I've seen a lot of use cases. We have helped one company, for example, you go to a grocery store. Today you ask for a grocery. Uh, rather than asking for grocery, you can ask for a whoa, I want to cook a recipe like I want to have pasta. From pasta you derive all the groceries you want to do and you place the order. So it becomes the unit of consumption changes. To me, the fundamental shift is you make the experience frictionless, you are more focused on the outcome that I am going to cook this and then the rest of the thing value comes through.

I think you can say the same thing for something on airline. If you wanted, like, I want to go to vacation and suggest to me a place and then book all the things that are relevant to do that. That's a new change from, from an airline or from a what I would say, Oh, book this airport and book this stuff. I am now changing. It's on its head. What do I do? We have seen both of them. 

I think there's a lot of change that can happen in training. In contact center, for example. You can actually empower end users to do themselves a lot of cases, which was always the case, but you can also show them like, okay, with audio visuals and all the things coming up, the multimodal that's coming up, you can show them and once you have a, like personally in my life, as I say, if I have to fix at home, I look at a YouTube video how to do that, and then I think if you can enterprise can do on a YouTube video on the fly, uh, something like that. I think that will be power. 

Saam: Ramesh, like maybe picking one of those examples, either the grocery example or the airline example, like, could you double click a bit and just talk us through how did the enterprise identify that that was an opportunity for AI. Like, what is the approach to actually building something, you know, pros and cons of, of, of trying to build your own versus working with a vendor. Like, let's say someone's listening and they're like, Oh, I have an idea for how this could transform my business, like, what's the next set of steps they should take?

Ramesh: I think there are obvious steps to do. One is, um, I think the fundamental question is this chat, the new interface. A lot of cases we should be asking that we share the new interface for everything. Google changed a lot of things and I used to chat the new. But beyond that, I would just say is, first and foremost, every organization needs to understand what are the key use cases across enterprise. And this is where it comes down to, what is my business, what is my disruption coming up being does this day in and day, day out? And we have helped thousands of clients to think about strategically, what are the use cases? What is your strategic direction, and what are the use cases in enterprise? That's to me a top thing to include in your data.

The second part is a simple exercise to do a two by two. A famous uh uh, two by two is to say what data is available and what use cases are there and kind of plot and what's the most valuable data and what you do. What you don't want to do is most complex use case up front because it's the engine is not warm up. We need to prove the value. 

So thinking about this, what data is available, what's the value at the stake, you can actually see which one to tackle because this is a multi year journey. So I would think about those two things and thinking that chat is the new interface that people are going to use to do that, I think you can unlock a lot of use cases in the process. 

Saam: If there's a CIO listening to this, like, are there two or three use cases you'd put in the simple horizontal bucket that, you know, if I was that CEO, you'd say, Hey, Saam, start with this. You can get it done. I feel confident you can show a win inside your organization and build momentum. 

Evan: Like the easy, like the easy wins, right? It's like, get started with these three easy wins. 

Ramesh: I think it depends on the context. Like if I was to start with, depends on whether you have depends on what conferencing tool you have. If you look at the obvious friction you can reduce, you can deploy Zoom AI or you can deploy Microsoft Teams. It is fantastic because everybody is taking notes and summarizing that and depends on the context. You can actually deploy any of them today. We have done that across the board. It has huge NPS for us.

So again, comes down to is there a friction to be removed? And I think that friction is to be removed because nobody wants to take notes. Nobody wants to consume that. I don't educate for one or the other. There are pros and cons, but I'm not going to get into that. But it is important to think about where we can reduce the friction.

The second one, I think the contact center is a use case for every organization for that, but it also comes down how it reaches your knowledge base. If your knowledge base is not rich, then the answer would not be as powerful. So you want to think about stuff. I think the question is not, do you have a perfection?

I think what I've seen from leaders sometimes will look streaming for perfection. There's nothing called perfection in the studio's world. What is important is a journey of iteration, what I call as MLP, most likable product. You continue to iterate on this because there is no perfect answer. The answers will change. 

I think the third important thing that I think every organization should do, which is I talked a lot of CIOs, is, is the architecture choices you make. Don't be bound by one versus other. You need to figure out a flexibility and architecture that enables you to change certain components on the fly and what are where are you signing up to do versus where you're not signing up to do. I would think about those three key things. 

Obviously, it is given that you need to look at risk and compliance because that is it. This technology doesn't come with its own sets of risk. It doesn't come with its own sets up. So you need to establish compliance. Proper risk and compliance forums. You need to make sure it is, uh, you are in lockstep with legal teams and our risk teams. And we have established some of the new forums because we want to do it the right way. We want to make sure we're training the right way. We want to change the right behavior. Like anybody using ChatGPT today, we asked them sign AI code of conduct because we it is a new way to operate and everybody should be understanding what we do.

Saam: Ramesh, you talk through a number of the use cases that, you know, you see clients using today, uh, and building today. Like, if you take a five year out view, and you think about kind of the technologies improvement, model scaling, continuing, multi modality, and the rest, When you, when you talk to some of the clients, like what are the, like, out there use cases people would like to do that you think are particularly exciting and you're bullish about?

Ramesh: I think the question is, are we always going to have a little bit of hallucination in the process? And how do you eliminate that? And if that is the case, Do you need always a copilot to check on the stuff? Maybe you get 80 percent accurate answer or 90 percent accurate answer. But is there a copilot cancer all the time to get to particularly certain critical things that cannot live with imperfections? It is not, uh, okay, if we give you a wrong hotel, the world is not ending. But if we give a wrong diagnosis to a human being, that's the wrong thing to do. So there may be certain use cases where we want 100 percent perfection and that has to be a human led. Human validated in some cases, and there may be other things where we can say, it's okay, we can, we can live with a little bit of imperfection because, uh, that there's not not. 

I think we have to get comfortable that there is a little bit of hallucination. In certain cases, we ought to understand that whether we, we want to live with that, or we want to not live with that. I think that's, that's a key question to be answered for. But I feel like technology, the models, the quality of the, what I call a summarization needs to get better. 

If you look at it from every model, I think they do good at aggregation, but quality still needs to get much, much better. And I know all the vendors are doing that. I think the security needs to get a little bit more better in terms of doing that and making sure we apply that. And I think that experience, if you think about that, even the whatever tool you do, I think the way to innovation Like, if you think about, uh, any of these tools, like whether it is Gemini or ChatGPT Enterprise and whatnot, how do you actually, kind of harness that power? We don't have that easy. I know every vendor is working through that to think about how do I expose to the enterprise, but how do we Even beyond enterprise, are there new use cases that we can actually expose to some? So to me, there's an app store type of, uh, play that is out there, which we can do today.

Evan: I personally struggle with thinking about the future about AI because I'm an entrepreneur, so I'm kind of an optimist where I hear all these things. I'm like, of course we could do all those things. It's going to be a basic thing. But I'm also like a software engineer, so I'm a little cynical where I'm like, ah, this stuff's not going to pan out. So I feel very, I have kind of two minds about the future.

Are there things you hear from your peers or maybe on the media where you say, I don't, that seems a little bit farfetched to me, right? Or any area where you feel a bit bullish on, you know, AI's opportunity for impact? 

Ramesh: I think it's a fantastic question. What I would say is my, my point of view is, see, AI can only do what it has, what has historically happened. Even if you assume everything was perfect. What AI cannot do, what is net new, completely. 

You can create a variations of what was there, yes, historically. Now you can assume how much of that is digitized versus not digitized. Whether history repeats itself or not, I think people say that it repeats, but like, so you can think about, we are, in a way, we are synthesizing the information. We are, uh, capturing the information, create new use cases, but we are not creating anything net new in a lot of cases. We are argumenting stuff. 

So I think it is, one can comprehend what cannot it do is something that new, which is what human brain should be thinking about. What, what is the experience we want to give? How do we think? So to me, there is an aspect of this that is still that we don't know what like net new things that we need to do because the innovation theoretically can go down because we're always kind of, doing things on the edge rather than completely redoing, re rethinking the different way.

I think the rethinking is what is the human brain should be doing rather than actually computer doing what has been historically done actually, and kind of, uh, creating variations of that. 

Saam: Yeah, I totally agree with that. Sorry, I'm going to jump in, but I just want to add one comment because I also, I think it also links back to what you're saying about hallucinations, Ramesh, and like one framing I've been using is there are use cases where hallucinations are a feature, and there's use cases where hallucinations are a bug, right?

And so in cases where I care about a very precise outcome, I would actually say we're far from, you know, precision and control in most of these use cases. And on the other hand, in use cases, and like the most obvious one is creative use cases, right? Whether it's, you know, First draft of writing, first draft of an application, first draft of a photo or video where I want creativity, like hallucination is some form of creativity and it actually is like a big feature. And so that's one other lens I've got. 

Ramesh: Yeah, absolutely. That makes sense. 

Evan: Ramesh, at the end of each episode we like to do, kind of a quick lightning round. So looking for like kind of the one tweet, you know, version of the answers and just, we acknowledge these questions are very difficult to answer in the one tweet format, so, you know, apologize in advance, but, um, Saam, you want to kick it off for us?

Saam: Yeah, so to start, Ramesh, how do you think companies should measure the success of a CIO? 

Ramesh: I think about three things, as I said, it's not one, top line growth, efficiency, value it creates and the cyber, I think about all three. 

Evan: Ramesh, you've been in this industry for a long time. You worked at a couple, you know, several different, like amazing leadership teams. What's maybe like one piece of advice you wish someone told you when you first became a CIO? Or maybe what advice would you share to up and coming CIOs? 

Ramesh: It's hard to say one. I think I say two, if you, if you allow me. One is, I think I said this poster, technology is now a disruptor. I think if you are, I, till now the success for every technology leader was keep things right and running, because that's what your definition was. Today, the success is how are you changing the business and how are you operating both? To me, blending strategy with execution is hard. Is hard, and that's a, that's a new muscle for every IT organization to do that. 

And I think if you ask me, second thing, what I learned from my own life is the relationships are critical and building relationship in an enterprise. It's so, uh, sometimes it's hard for tech folks because, and I come from my own stuff. We think that we're doing all the stuff. That, that to me, I would say is a lot, uh, two things that come to the top. We can keep on going for other stuff, but those two things is relationships and a blending strategy with operations.

Evan: How do you think CIO should best position themselves across their colleagues, right? In order to, you know, maximize the opportunity to maximize relationships, to help them have the biggest impact? 

Ramesh: I think it's important to kind of be to understand. I think about, uh, I created my own version of what I call is like the relationship map. I think one important understanding is who is my key stakeholder and who am I going to connect with? And what is that I'm going to connect him him or her to? What am I doing? So I think about these four key stakeholders that most people I think five now five stakeholders that you should manage. 

One is I call up and down. You should manage up. You should manage down. Should manage left. You should manage right. Your stakeholders, your customers, as well as your peers. And then you should manage your brand outside to make sure you people know you. So when I think about that, what should we do is to look at all the stakeholders and balance your time between execution and strategy to say, am I connecting to the right people 'cause you might be doing all the right things, but nobody knows that you, unfortunately, most IT organizations do not have a marketing department. So then you don't know. People don't know what you do. So you are kind of, perception is the reality. Whether that's a real thing or not. You, you have to create your own perception.

Perception or a new thing that actually people, people realize what you do. 

Saam: Maybe switching gears to the more personal side, um, Ramesh, what's a book you've read recently that's had a big impact on you and why? 

Ramesh: I'm a big believer in servant leadership mindset. I recently am reading the book from Daniel Goldman called Optimal, The Optimal. He just released it. I follow him a lot. So I think, uh, that's a book I recommend to everybody is, is fantastic. 

Evan: And final question for me in the episode, what do you think will be true about technology's future impact in the world that most people would consider science fiction today? And I'm looking for kind of like your contrarian take on, you know, what the future can be like that, you know, maybe most people wouldn't agree with? 

Ramesh: I think there's a positive side of it. There is a negative side of it. I mean, the sustainability area also. The positive side of it is I think technology is to a point where it can deploy, uh, deploy, differentiate, deliver incredible amount of value. I, there's no doubt about that and I think it's proven to do that. 

The downside of that, even if you look at AI, today, us power grid was stable for the last 20 years and whatnot. More and more, it's becoming critical that AI itself can generate so much. Can deliver so much power that we don't have. So we have, we have a, we're at a conundrum in terms of the power consumption. How do you solve this problem? Cause we might have the technology, but we might not have enough resources to deal with it. And so that's going to be a humanity challenge to me. How do you, how do we, how do we navigate this ecosystem? 

Evan: Ramesh, thank you so much for joining us today. Really enjoy the conversation and I'm looking forward to chatting again soon.

Saam: Thanks a lot, Ramesh. 

Ramesh: It was wonderful to have a general conversation.

Evan: That was Ramesh Razdan, Global Chief Technology & Information Officer at Bain & Company.

‍Saam: Thanks for listening to the Enterprise Software Innovators podcast. I’m Saam Motamedi, a general partner at Greylock Partners.

‍Evan: And I’m Evan Reiser, the CEO and founder of Abnormal Security. Please be sure to subscribe, so you never miss an episode. You can find more great lessons from technology leaders and other enterprise software experts at enterprisesoftware.blog.

‍Saam: This show is produced by Luke Reiser and Josh Meer. See you next time!