How IBM is Harnessing Gen AI for Client Success

Brian Bohan (29:58):
Hello and welcome to the Executive Insights podcast. I'm Brian Bohan, I lead the consulting partner Center of Excellence at AWS, and today I'm joined by Mahmoud Elmashni, who is managing partner at IBM Consulting and leads the AWS Alliance there as well. Mahmoud, thanks for joining me today. Really excited to talk to you.

Mahmoud Elmashni (30:17):
Thank you for having me. I'm excited to be here.

Brian Bohan (06:22):    
And so if you could tell us a little bit about what you do at IBM Consulting, your role, your responsibilities?

Mahmoud Elmashni (06:27):
So I run our AWS business globally. And really it's been one of our fastest growing business overall with one of our partnerships. Now, we're over 25,000 certified people just on AWS. The number of competencies, the number of joint client stories we've published, everything. It's really been a rocket ship over the past couple of years, so it's been really cool to be a part of.

Brian Bohan (07:03):
So the first thing I wanted to touch on is generative AI, and we've talked about this before, really has the potential and the power to truly transform business processes, applications end-to-end businesses in their entirety. Really interested what you're seeing with IBM Consulting, both in terms of how it's changing the practice and how the organization itself is interacting with clients. And then what are you doing with your clients with gen AI?

Mahmoud Elmashni (07:27):
So look, with gen AI specifically, and I actually key off a little bit of something that I heard Werner Vogels in Tokyo months ago during that summit. And he said it's going to get to a point where gen AI is a technology just like any other technology. So we don't talk about the technology behind a microwave oven, we heat our food.

(07:52):
We don't talk about technology behind the fiber optic cables, we just use the internet. So for us, what we're seeing is that, instead of it being more about digital everywhere, it's about gen AI, here and now. And when it comes to a consultancy, I think the biggest thing that we've seen is that it's not about gen AI replacing an individual. It's about supplementing the individual and making them that much more productive to be able to execute.

(08:21):
So that's really where we've focused on a business, which is if you're trying to scale your business and trying to take on these gnarlier and gnarlier transformations for your clients, how do you enable your practitioners, and your entire organization to be able to execute with gen AI? You introduce these assistants to make them that much more productive to be able to deliver for your clients. So for us, we simplified it a bit and introduced the IBM Consulting Advantage platform so that this way they can have access to all of those assistants while they do their everyday work.

Brian Bohan (08:56):
And I love that you said that going forward we're going to talk less about a gen AI project and more just about what we do every day, helping our clients and customers transform their businesses, get value, and some of that is going to be powered by gen AI. Absolutely. And have you seen it just in terms of the partnership itself and how IBM consulting and AWS are working together? Any changes or impacts there?

Mahmoud Elmashni (09:21):
Look for this, we've really kind of focused our business on what we call the science of consulting. It's around the four P's, makes it easier to remember, right? The people, process, partnerships and platform. So you hit on, if you briefly look at the people that's making sure that they can have access to what they need to be able to be more productive. The process, everybody's running around trying to do POC purgatory as I like to jokingly call it, until they can figure out what they want to do and not want to do. The partnership aspect in coming in, that's been absolutely key for us in terms of being able to scale.

(10:03)
So I mentioned the IBM Consulting Advantage, which for us is the platform. The reality is that you can access HAQM Q and all of the stuff that now AWS and HAQM are announcing left, right, and center around gen AI for our practitioners to be able to deliver for our clients to make their lives easier. So it gave us that one platform, but still to be able to access all of the AWS technology. So we've invested heavily in training our resources on being able to use it, understand it, know when to call it, and know when to use what pieces and everything to be able to deliver for our clients.

Brian Bohan (25:30):
That’s great. This gets me super excited about the partnership, a big part of our partnership is also helping our customers and clients move out of their data centers and modernize their applications and workloads. And I think one thing that's so exciting about gen AI in the past there's always been this trade-off. You can lift and shift then modernize. And if you modernize while you're migrating, it might extenuate at your timeframes or increase your costs. And I think now with Q Transform, and gen AI, we can have both. We can migrate and modernize simultaneously while keeping those timelines and costs really fixed like they used to be. And I'm just curious what you're seeing with your clients with IBM in terms of how you're helping them apply gen AI to those modernizations?

Mahmoud Elmashni (26:14):
Yeah, look, if you're CIO and you're going to put forward a 10-year road map to modernize, that's pretty much a kiss of death. I mean, so the reality is that gen AI has allowed us to take a look across a number of technologies to be able to modernize them in a much faster period of time. You're talking now six, to 12, to 18 months. No longer multi-year. What can I get done immediately? So that could be applied to mainframe modernization, could be applied to VMware modernization. It could be applied to, as you mentioned, I'm looking at these big DCs and thinking it's like, okay, how do I exit out of this data center in some kind of timely fashion that's not going to carry on to the next generation.

(27:07)
And even some things as simple as upgrading Java code for a client is happening in months and not years anymore.

Brian Bohan (27:35):
Absolutely. That's fantastic. Yeah, we're seeing the same thing and we're really excited about just in the faster we can get our customers into the cloud, modernize, the sooner they'll be seeing the value.

Mahmoud Elmashni (27:45):
And I think also the return on investment for themselves, right? Because they're all looking at top line revenue growth and how do I take these dollars and reinvest it back into my business? And not spending a fortune in dealing with this technical debt that's wrapped around my neck? Right? So-

Brian Bohan (28:00):
Yeah, absolutely.

Mahmoud Elmashni (28:00):
... I think that's one of the big things that we can bring to bear for a lot of our joint clients.

Brian Bohan (10:58):
So if you talk a little bit further about people, you mentioned the four P's and obviously IBM Consulting is very much a people business. And at the end of the day it's people working with your customers to get things done. So again, touch on a little bit more about the skills and the training and then even both the technical level and the executive level and how you're addressing that.

Mahmoud Elmashni (11:22):
So we've really, when we first started to embark on the gen AI journey and the training associated with it, we wanted to make sure everyone in the organization had some level of understanding. It's not just HAQM Q for Developers, it's not just the developers. Everybody needs to basically have a fundamental understanding of the platforms, the various LLMs, how to access different things, et cetera.

(11:50)
So we laid out training that went from our managing partners, the very senior level of our organization, down to your entry level developer all the way through. Fortunately, we've used gen AI to help customize that. We developed specific curriculums for each individual. So that they know how to be able to have that conversation and deliver for their client. Simultaneously, we even started to customize that a bit. So last year, right around this time, AWS announced their training for a lot of their partners and everything. So we took that and built it into the AWS gen AI version of our training and deployed that. And now we're taking it a step further because AWS earlier this year announced even their first two gen AI certifications right around July timeframe or so. Shocker. They were data-based, right? I mean, so data seems to be the other common theme running around nowadays and how to use that with gen AI.

(12:53)
And so for us, we've invested heavily in terms of getting our people that sort of knowledge, whether it's to go deep and get certified, or even to just understand the general AI track and getting that AWS training to go along with that to be able to execute. And we've used gen AI even internally to help streamline some of that and deploy it that much faster to our personnel.

Brian Bohan (13:18):
Love that. And so you can take our more general gen AI technical training, but then especially as you apply it to the managing partners who have that deep domain expertise, industry expertise, they're able to contextualize it then for their clients.

Mahmoud Elmashni (13:30):
Absolutely. It's 160,000 person organization. So literally the plan was how do you touch 160,000 people?

Brian Bohan (13:38):
It's a big challenge.

Mahmoud Elmashni (13:38):
Yeah, absolutely. It wasn't just the developers, or just the partners, or just sales training. It was up and down the stack.

Brian Bohan (13:45):
Fantastic. So one thing, shifting a little bit to another P that you mentioned around process, and this is the process around value creation. And one thing that we're seeing certainly as we hopefully are moving away from the POC purgatory that you rightly call out. And we saw we're seeing increasing customers come out and talk about real value that they're deriving from generative AI. We're also seeing the conversation shift from just concerns around cost, to true ROI conversations. What is going to be the business value driven based on my investment? I'm curious how IBM is helping your clients really think about that ROI and value calculator around gen AI?

Mahmoud Elmashni (14:24):
Yeah, so, the POCs, you can run a lot of them, but they've taught us a lot and they've taught our clients a lot.

(14:32)
In terms of even knowing what to ask for and what not to ask for, right? So when you look at the process aspect, that means taking the entire end-to-end value of what that ROI is supposed to look like for the client. It's not about, okay, did this work for this one spot? Yes or no? Yeah, it worked fantastic. It was great. I got the results I wanted, but it cost me a small fortune with the LLMs I'm using.

(15:01)
So you want to factor that in, in fact, one of the jokes I'd heard before is LLMs could stand for losing lots of money if you do it wrong. So we want to make sure that we take that end-end approach, make sure that yes, you get the results you want, but was it the most efficient way to be able to get there? And by the way, not everything needs to necessarily have the technology applied to it too. So we're seeing pockets of that. If this works the way it's supposed to or you're going to sunset that app, why are you going to go through and invest all of this sort of refactoring effort just to do that now?

(15:37)
So it's trying to take that holistic approach across the board and the POCs gave us a good foundational base for knowing how to look at that ROI for a client. At the same time, our clients learned off of what to ask for and not to ask for now, for some of these POCs, right? I need to better understand, okay, what models do I really want to use? I need to better understand what platform do I access it through? So for example, Bedrock as a thing. So that's what we're finding throughout in doing this full ROI evaluation for a lot of our clients. And our clients are asking us when they retain a consultancy or an SI, they want an opinion. I mean, I want to see it work, but what do you think? Where do you think it's going to land me and show me that roadmap over the next 12 to 24 months?

Brian Bohan (16:27):
Absolutely. I mean, this is one of the reasons why we talk so much about selection and choice because we're so early in this journey and there's so many different things that we could be doing with this technology. So we want to make sure we have the right chips for your price performance, the right models based on the right use case, but with a lot of choice also comes a lot of choice.

Brian Bohan (17:26):
And so talking about the customers and the clients, and again, you all have very deep industry expertise as well, just interested in what you're seeing? How they're adopting gen AI and working with IBM Consulting to really make it relevant for their businesses?

Mahmoud Elmashni (17:42):
So probably the most common use cases that we see are contact center intelligence. Shocker. Image, document image processing, and being able to summarize, and that sort of thing. Application, migration, modernization, governance, security, right? If I really had to call out, those are the ones that we're seeing all over the place. And it's no surprise that everybody's announcements are all centered around those five.
    
(18:15)
And you're seeing it span a number of industries, right? Everything from life sciences, to healthcare, to financial services, to public. I thought, I didn't think the public sector would be so quick to jump on, but shocker, they are with everyone else too. Everybody's looking for try to get more efficient or do more with less, right? And telco and automotive. And so we're trying to make sure that now we take that approach that is industry specific. We've always gone to market by industry. So you look at taking that gen AI flavor and putting that industry flavor on top of the gen AI technical cap ability, to be able to deliver for your clients.

(18:59)
So taking those use cases then and applying those by industry to make sure that you can deliver. A simple example is, I mentioned public, we're working with a government agency and using gen AI to help translate code into modern Java code. Essentially compared to manual hours at the speed of light. I mean, it's insane how much you can save by just doing that and running the checks in the background as opposed to having manual conversion of bringing them up to speed in that regard. So for them down the road, it becomes much more efficient. They get cost savings off of less hardware to be able to run it. So it's a win-win. Not only in terms of speed of execution, but the long-term savings for them.

Brian Bohan (19:44):
That's a great example of a low hanging fruit to show immediate value. HAQM itself has saved, I think it's 4,500 developer years by upgrading all of its Java applications to Java 17.

Mahmoud Elmashni (20:02):
Absolutely. And that's what you're going to see more and more of I think as you go forward, right?

Brian Bohan (20:08):
So Mahmoud, thanks so much for joining me today. I'm really excited by the progress and the potential of our partnership with IBM Consulting. Super clear already that we're delivering some big results around gen AI for our joint customers. So again, thank you very much and really looking forward to what we do in 2025 together.

Mahmoud Elmashni (20:25):
Absolutely. Thank you for having me, and I'm excited about the future as well.

Listen now

Listen now

Listen now