State Street tech chief: ‘Blockchain is one of my biggest investment areas’

2 Mar 2018

State Street CTO Moiz Kohari. Image: State Street

State Street chief technology architect Moiz Kohari explains that investments in digital real-time latency, data science and blockchain are at the top of his agenda.

Prior to joining State Street in 2016, Moiz Kohari was executive vice-president and head of global technology innovation at the London Stock Exchange Group.

Kohari is responsible for blockchain, cognitive computing and low-latency initiatives across all State Street companies.

‘We are talking about data in hundreds of petabytes, which is reasonably significant and, as you can imagine, some of the most critical data on the planet’

Future Human

This includes driving advance research for the entire trading life cycle, starting from pre-trade and trade components to post-trade technologies related to clearing and settlement.

From 2005 to 2011, he worked with Novell as vice-president of engineering and then as chief technology architect.

He was also the founder and CEO of Mission Critical Linux.

State Street, founded in 1792, is the second-oldest financial institution in the US. It employs more than 33,000 people worldwide and has more than $2.4trn under management and $28trn under custody and administration.

Kohari was in Cork recently to speak at a fintech seminar held at the State Street Advanced Technology Centre. In 2015 State Street linked up with UCC to establish an R&D facility in partnership with Zhejiang University in eastern China.

Can you outline the breadth and scope of the technology roll-out across your organisation and what improvements it will bring to the company?

We run a significant portion of the financial business from a back-office, accounting services perspective, and most of these systems were created years ago running on top of legacy, proprietary infrastructure on top of mainframes, and systems like that.

This is no different from almost any other financial institution on the planet.

As we look at that, we find that, given the richness of that data we have available – the type of transactional overall capabilities that move through State Street, which are very unique to us – we find that leveraging and driving intelligence and implications with regards to market movements and other things like that, it is very difficult to drive that level of intelligence from the systems that we have present today.

This is because these systems are operating in different silos, and the data is in multiple silos. So, in order to get a holistic view of that data, you have to really propagate through many, many different systems and, as you start to do that, the challenge of pulling that data together to be able to analyse it is extremely large.

And it puts a lot of stress on the system, so you are unable to actually do the type of analytics that you want to do.

So, in order to evolve our organisation to be able to compete in this new digital world, we believe it is absolutely critical and necessary to update and enhance our platforms to be able to take advantage of open platform technologies; all the way from the hardware that you run the operating systems on top of to creating a microservices-based architecture, which allows you to break down all of the different services you provide in a granular fashion.

So, you can now start to take these services and create macroservices off of those microservices.

By doing that, you are creating reusable architecture.

With these reusable components, you can pretty much create any type of result that you want.

These components have to be created, in our opinion, on top of modern technologies leveraging open source and things like that, so that it is actually relatively easier to hire talent that can work on it instead of talent that will work on some of the proprietary technologies that are available in the marketplace.

How big is the scale of the challenge?

If you think about it, we are talking about data in hundreds of petabytes, which is reasonably significant and, as you can imagine, some of the most critical data on the planet.

The challenge is correlating it with real-time, unstructured, streaming data from hundreds of different sites on the planet and then curating that data and then providing insights to our clients.

And really, if you think about it, outside of the fact that we are a trust bank and provide trust and custody services to our clients, the true benefit that we bring to our clients is that data and the insights we have.

We are literally recreating the entire platform to do just what I painted for you.

With fintech rising and cryptocurrencies all the rage, what is State Street’s approach to blockchain technology?

One of my three largest investments in technology is around blockchain.

In terms of cryptocurrency, at least at the moment, the bank has no defined plans around it and I am personally not spending much time on cryptocurrency.

I am spending a significant amount of my own time and my team’s effort in creating and managing blockchain technologies. So, to that effect, we analysed almost every single blockchain technology out there over the course of the last four to five years, and we were very early in engaging in this space.

As a result of that, we have actually honed in on what we want to be able to leverage and, now that we’ve done that, we have a significant portion of our team that are not only contributors, but maintainers of these projects in the open source community.

Some of the most relevant and well-known entities in this space today work with State Street and for State Street, and we are creating some of our core back-end capabilities on top of these technologies, which we will start to bring to market over the course of the next 12 to 24 months.

And, as we start to do that, you will start to see an evolutionary effect on the types of services we will be able to offer to our clients.

Because, if our clients are operating on top of these capabilities with us, they will have real-time information and capability of real-time exchange with us on top of that platform.

In short, there is a very significant investment that we are making in this space. But, to be very clear, it is completely different from anything that the current cryptocurrency market does and which operates in an anonymous network.

What we are doing is creating and operating on top of a known network of participants.

As banking becomes more data-centric, how should people be preparing themselves for the jobs of the future?

Data science, in my opinion and in my lifetime, is not going to go out of style.

Data science is all about mathematics, statistical and probabilistic theories.

Now, data science, as we’re applying it in our environment today, is from a cognitive computing perspective; how you create capabilities around anomaly detection, around predictive analytics, around creating different types of natural language processing capabilities.

We are leveraging it all across the board.

Another core area, if you are talking about where we would like to see more computer scientists concentrate, is around infrastructure; how you take ingestion capabilities and take them into a storage architecture, and on top of that create appropriate ontologies and lineage capabilities, so you know actually what is in that storage architecture.

These are extremely critical concepts and you have to be able to understand computer science and engineering at the lowest levels in order to do a good job of that, because what you are talking about is your ability to create systems that perform at real-time speeds so that information exchange is real-time and the entire world is moving towards that near-real-time capability.

In order to do all of that, the basic concepts are absolutely critical and, where we concentrate on evolutionary and technology, point of view is absolutely critical.

So, cognitive computing, data sciences, infrastructure capabilities and data modelling capabilities, and metadata creation – these are core areas of development where I literally have hundreds of engineers working, and I’m looking to hire hundreds and hundreds more.

Want stories like this and more direct to your inbox? Sign up for Tech Trends, Silicon Republic’s weekly digest of need-to-know tech news.

John Kennedy is a journalist who served as editor of Silicon Republic for 17 years