The Tug-Of-War Between Tomorrow’s AI and Yesterday’s Legacy Systems

For as long as people in any technology field can remember, there’s always been a sort of push and pull, back and forth between the latest technological innovation that may change the world as we know it and the legacy systems we have to preserve for a while longer.

Old client server systems were relatively large-scale programs. Then they got replaced during the next wave with microservices architecture. One program would be divvied up into small pieces that would have microservices and those services could be used in a variety of applications. 

The new wave we see today with AI and machine learning represents a completely different type of architecture. Rather than an entity built on microservices, AI has algorithms that are inverting matrices over and over.

As a result, machine learning delivers excellent potential that is already being incorporated more now in systems than just a couple of years ago. Still, some firms have trouble moving forward full speed into the future when they have to keep the legacy systems part of the house up.

What’s the solution here?

Most of the companies we’ve done business with at Roy Talman & Associates haven’t been using 40-year-old technology today, but we do still hear stories about people saying, “OK, so we’ve got a 15-year-old system. It still works, but we’re always worried that it’s going to stop working one day. Does that mean we should go out and build a new system? Possibly but that’s also a costly proposition.” From the management’s point of view, it’s a tricky decision. Some companies hope that as they build a new system, it will essentially overtake the old system anyway, so they will not need to go to Version 3.0 of the system anytime soon.

Going “All In” On The Future

Rather than trying to prop up a legacy system, a firm could say, “You know what? Let these people who understand the decades-old system retire. We’re going to be looking forward toward the next system and keeping the present-day system in place – but even that will be newer.”

This happens quite often. One banking giant has had one system that took seven or eight years to build for $500 million. As they look ahead, a replacement system costs about $700 million. Yet, as the new system is being built, they cannot retire the whole of the prior system at once. So parallel systems are being run.

Slowly but surely, in time, the new system overtakes the old system. But guess what happens? By the time the new system takes over, it’s seven years old. So the bank is building the next generation of the system, which may cost up to a billion dollars.

This process repeats in which every round of implementing a new system takes 7+ years. By the time that system is fully functional, it’s time to replace it. The long cycle of building and replacing can feel foreign for many, primarily since we’re used to quickly updating apps on our phones.


Talman Advantage #7: We Already Know Many People At The Top

The built-in advantage of being a more specialized recruiter for over three decades is that Roy Talman & Associates established many strong relationships with senior leaders in the C-suite and Director level. How do we truly leverage that? Prior to your interview, we can provide you terrific insight on the person’s background, the questions they’re likely to ask you and even a few clues into why prior candidates were likely rejected. A recruiter that equips you with more information in advance of the interview? That just might make all the difference – if you talk to Talman first.


Still, companies are realizing that the model they need to adopt requires a different architecture for the system. Enter machine learning.

Andrea de Santis for Unsplash

The pace of change and adoption of machine learning is ever-accelerating. Take Elon Musk’s Dojo – a chip recently revealed by the innovator specifically designed for machine learning applications. Musk is also promising to build a Dojo computer that will be specifically for machine learning and targeting image processing applications in particular.

The technology and high-frequency trading companies we work with at Roy Talman & Associates embrace a constantly evolving mindset. There are always new iterations moving toward the best of the best technology and the best of the best talent.

The ones who are not taking that approach are going to be the ones quickly falling behind.

Take the banking industry as an example of the danger of hanging onto a legacy system too long. You would assume many of the large banks would have a very sophisticated technology setup. In some areas, that’s true. Yet, several are hanging on to their 30-year-old mainframes. Is there any problem with that right now? No. For all intent and purposes, so far, so good. However, what happens as time goes on? These financial institutions may be in for a rude awakening the longer they deal with old mainframes for a “good enough” way of doing business.

Companies from PayPal to Coinbase haven’t been standing still on legacy systems. They have eyes straight ahead on what’s next and their market cap reflects that. It only seems to some on the outside looking in as if they created massive value “out of thin air.”

The longer a company hangs onto its legacy technologies with the rationale of “it still works,” the shorter the runway they will give themselves to catch up to the competition or get passed by others. 

In our next Tidbit, we’ll document a different kind of race of digital transformation going on between names you know like Intel, Google, Apple and Microsoft that will shape the way companies design and manufacture chips going forward. And with a select number of companies with these particular resources in the world, the implications for the rest of us are huge, impacting everything from the cars we drive to the computers we use.

Let’s face it. Some companies are caught in a tug-of-war between hiring technical talent that understands the legacy system you’ve had in place for many years and hiring technical talent that’s made for what’s coming next. Yours might be one of them.

So before you make any sudden decisions, Talk To Talman First. A conversation with our team at Roy Talman & Associates about where you are today and where you want to be can help shape a hiring strategy that works with your growth rather than holding you back. It’s why we’re the technical recruiter that companies consistently count on to identify and bring aboard the “top 1% of the top 1%.” Now it’s time for you to experience it for yourself.